Chest Physiotherapy in the Intensive Care Unit, 2nd Edition

Chest Physiotherapy in the Intensive Care Unit SECOND EDITION

Physiotherapy in the Intensive Care Unit SECOND EDITION

Colin F. Mackenzie, M.B., Ch.B., Editor Associate ProCessor Department of Anesthesiology University of Maryland School of Medicine Baltimore, Maryland

P. Cristina Imle, M.S., P.T. Physical Therapist Maryland Institute for Emergency Medical Services Systems Clinical Instructor. Department of Physical Therapy University of Maryland School of Medicine Baltimore. Maryland

Nancy Ciesla, B.S., P.T. Chief Physical Therapist Maryland Institute (or Emergency Medical Services Systems and Montebello Rehabilitation Hospital Clinical Instructor. Department of Physical Therapy Universily of Maryland School of Medicine Baltimore. Maryland

With a contribution in the 1st Edition to Chapters 5 and 6 from Nancy Klemic, B.S., P. T. Formerly, Senior Physical Therapist Maryland Institute for Emergency Medical Services Systems Ballimore, Maryland

WILLIAMS & WILKINS Baltimore' Hong Kong' London' Sydney

Editor: Timothy H. Grayson Associate Editor: Carol Eckhart

Copy Edilor: Arline Keithe. Amy Redmon Design: JoAnne Janowiak Illustralion Planning: Lorraine Wrzosek Production: Raymond E. Reter

Copyright

1989

Williams & Wilkins 428 East Presion Sireet Baltimore. Maryland 21202. USA

All rights reserved. This book is protected by copyright. No part of this book may be reproduced in any form or by any means. including photocopying. or utilized by any informalion storage and retrieval system without wrillen permission from the copyright owner. Accurate indications. adverse reactions. and dosage schedules for drugs are provided in this book. but it is possible that they may change. The reader is urged to review the package information data of the manufacturers of the medications mentioned. Printed in the United SIOles of America First Edition 1981 Library of Congress Cataloging-in-Publication Data Chest physiotherapy in the intensive care unit/Colin f'. Mackenzie. editor: P. Cristina Imle. Nancy Ciesla. Nancy Klemic.-2nd ed. p.

em.

Includes bibliographies and index. ISBN 0·683·05329·9

1. Lungs-Diseases-Physical therapy. 3. Critical care medicine. Christina. (DNLM:

III. Ciesla. Nancy.

RC735.P58C48

II. Imle. P.

IV. Klemic. Nancy.

1. Intensive Care Units.

3. Physical Therapy.

2. R espiratory therapy.

I. Mackenzie. Colin f'.

2. Lung Diseases-therapy.

WF 145 C526)

1988

617'.5406-dc 19 DNLM/DLC for Library of Congress

88·37426 CIP 90 91 92 2

3

4

5

6

7

8

9

10

Preface to the Second Edition change pulmonary pressures and assist in the clearance of retained secretions. Chapter 2 provides new sections on res­ piratory physiology and respiratory mucus. Scientific data to justify postural drainage, positioning and breathing ex­ ercises, and percussion and vibration are updated in Chapters 3 and 4. Chapter 5 compares the effects of coughing and forced expiration and reviews the cur­ rent techniques recommended for airway suctioning. Chapter 6 summarizes the important new aspects of patient mobili­ zation used to minimize the need for chest physiotherapy. Chapter 7 includes more information on collateral airways and synthesizes some postulated mecha­ nisms of action of chest physiotherapy. Treatment of patients with brain and spine injuries is comprehensively up­ dated in Chapter 8. Chapter 9 covers some newer adjuncts to chest physio­ therapy, most notably positive pressure techniques. Pain management tech­ niques in the ICU are expanded in the re­ vised Chapter 10. Statistics of patients treated since 1974 are now consolidated in Appendix I. Abbreviations and sym­ bols are defined in Appendix II. An up­ dated summary of chest physiotherapy treatment and evaluation is presented in Appendix 1II, and Appendix IV details the interventions in four critically ill pa­ tients. The index is now more com­ prehensive.

The second edition of Chest Physiother­ apy in the Intensive Care Unit follows 8 years after the first. The object remains to provide a comprehensive reference source for all professionals involved in respiratory intensive care. Over 900 references are provided. A new section summarizes respiratory physiology and there are comprehensive reviews of res­ piratory mucus, cough, and suctioning techniques. Every chapter is extensively updated. This new edition adds over 300 references from material published since

1980.

The second edition addresses contro­ versies about use of postural drainage, percussion, vibration, breathing exer­ cises, cough, suctioning, and mobiliza­ tion. Current theories and conflicts about indications and contraindications for chest physiotherapy are described and discussed. The authors' 15 years of ex­ perience of managing chest physiother­ apy for critically ill patients is presented. The format of the second edition remains unchanged because the first edition had three printings, was published in Spanish and Portuguese and was favorably re­ viewed. In Chapter 1 of the second edition, crit­ ical summaries of literature since 1981 describe pathophysiology of respiratory complications and techniques including chest physiotherapy, bronchoscopy and positive pressure, all of which are used to

v

Preface to the First Edition remarkable record of service. The three physical therapists who have contributed to this book have had betwee!l them 18 years of work at M[EMSS since 1973. The therapy they have provided encompasses five maneuvers: 1) postural drainage, 2) chest wall percussion and vibration, 3) coughing, 4) suctioning of the loosened secretions, and 5) breathing exercises in the spontaneously breathing patient. [n addition, mobilization is used whenever possible. Besides the similarities in patient pop­ ulation, personnel, and therapy, the me­ chanical ventilatory support was stan­ dardized at the Institute between 1973 and 1978 with the use of only one type of volume-present ventilator. Controlled mechanical ventilatory support was em­ ployed for resuscitation, for anesthesia and throughout recovery, providing hu­ midification at all times. From October 1978 on, intermittent mandatory ventila­ tion was occasionally used instead of controlled mechanical ventilation. No in­ termittent positive pressure breathing (IPPB) machines were used to deliver bronchodilator or mucolytic agents. No inhaled drugs. other than water vapor, were given in the critical or intensive care units. Tracheal lavage was rarely employed. The "bag squeezing" tech­ nique of chest physiotherapy, in which the lung is hyperinflated and the chest vibrated during expiration, was not used. No spontaneously breathing patients were treated with the aid of blow bottles or incentive inspiratory spirometers. Na­ sotracheal suctioning was seldom used or attempted. Tracheal suctioning was only carried out in intubated patients. Be­ cause these other respiratory maneuvers were excluded, the effect of chest phys­ iotherapy alone was determined. As with any book directed at diverse groups, such as critical care specialists,

It is quite apparent, from even a casual conversation with physicians or other personnel involved in respiratory care management, that there is a large spec­ trum of differing treatments termed by their users as "chest physiotherapy." The literature is not helpful in specifying what chest physiotherapy is intended to include. [s the inhalation of bronchodi­ lating or mucolytic agents part of chest physiotherapy? In tracheally intubated patients, is manual hyperinflation of the lung an inclusive part of chest physio­ therapy? Many centers would use these therapies, others would not. All would claim to be treating the patient with chest physiotherapy. It is not surprising, there­ fore, that there are many contradictory opinions concerning the effects of chest physiotherapy. Because of these varia­ tions, if an improvement does occur, it is likely to be difficult to determine the beneficial component. A homogeneous patient population treated in a similar manner by the same personnel over a number of years gives a useful clinical experience that fre­ quently cannot be duplicated. At the Maryland Institute for Emergency Medi­ cal Services Systems (MIEMSS) in Balti­ more, Maryland, there is a unique and homogeneous population of traumatized patients. Year after year, the admission statistics confirm the similarities in the patients, and their injuries, and in mor­ bidity and mortality. The population is unique because about 60% of the 1,200 or more patients admitted each year come directly from the scene of their accident and about 75% of these patients come to the Institute by helicopter. For the first 7 years, chest physiother­ apy has been used in the critical care and intensive care units to treat patients with lung secretion retention. The physical therapists providing the therapy have a vii

viii

anesthesiologists, surgeons, internists, chest physiotherapists, nurse intensiv­ ists, and respiratory therapists, some areas of the text are more relevant than others to each group, For the physician, the changes that take place with therapy and the aggressive approach taken at MIEMSS are complemented by a consid­ erable quantity of data and many case histories, To the physical therapist work­ ing in the intensive care unit, this book provides complete coverage of the spe­ cialty of chest physiotherapy, For the nurse intensivist and respiratory thera­ pist, a practical approach to the respira­ tory management of the multiply-moni­ tored intensive care unit patient is combined with a reference guide to the literature on chest physiotherapy, This

PREFACE TO THE FIRST EDITION

book presents our experience with chest physiotherapy in the management of acute lung pathology in patients with previously normal and abnormal lungs, Over the 7 years (1973-80), a homoge­ neous patient population of over 3,000 in­ tensive care unit patients was treated. The mechanical ventilation and physio­ therapy techniques were standardized, and the medical and physical therapy personnel managing the respiratory care were constant. It is hoped that this book will provide others with a well-tested, practical approach to chest physiother­ apy for intensive care patients. C.F.M. February 1981

Acknowledgments

First Edition. The majority of this book was written from knowledge acquired at the Maryland Institute for Emergency Medicine under the direction of R. Adams Cowley, M.D. We are greatly in­ debted to our mentors and colleagues, who have worked at the Institute with us during these years, for their teaching and assistance. Particular acknowledgment must be made to T. Crawford McAslan, M.D., who was Clinical Associate Direc­ tor of MIEMSS. Under his guidance, chest physiotherapy was introduced to MIEMSS in 1973. He and Baekhyo Shin, M.D., provided a stimulating intellectual environment in which to study clinical respiratory physiology and the effects of chest physiotherapy. Our debt to Drs. McAslan, Shin, and Cowley is very great. We thank our colleagues, physicians and nurses. for their help and cooperation with the production of this book, and Gareth Green, M.D., Editor of the Amer­ ican Review of Respiratory Diseases, who kindly supplied drafts of the November 1980 supplement. We also owe thanks to Mark Moody, Ph.D.. Director of Clinic and Field Evaluation at MIEMSS, for data concerning admissions appearing in Ta­ bles 1.1-1.6, and to T. Crawford Mc­ Asian, M.D., for the traces appearing in Figures 1.2-1.4. For the photographs, we thank Colin Mackenzie, M.B., Ch.B., F.F.A.R.C.S., and Dick Register for taking them; and Frank Ciesla, MIEMSS and University of Mary­ land Hospital Illustrative Services, for printing them. For illustrations, we thank Chris McCullough-Green; and for proof­ reading. Barbara Eerligh and Beverley Sopp. Jeremy Hallisey, M.B., B.S., and David Clark helped with data analysis, proofreading and reference checking. Our thanks go to Marlene Wheeler and Kate McWilliams, who typed the final

drafts, and to Sandy Bond-Lillicropp, who organized the typing of the earlier drafts. Finally, we thank experts in the field of chest physiotherapy on both sides of the Atlantic who have read and criticized the manuscript at various stages. However, the final result should not be blamed on them. Rather, the end product is the re­ sult of our determination to keep some parts, such as the patient population data and the sections on special patients and mobilization which do not relate strictly to "chest therapy" or intensive care." The reviewers included Margaret Branthwaite, M.R.C.P., F.F.A.R.C.S., and Barbara Webber of the Brompton Hospi­ tal, London. England; Anthony Clement, M.B .. B.S.. F.F.A.R.C.S., of St. Thomas' Hospital, London; John Hedley-Whyte, M.D., and Cynthia Zadai of Beth Israel Hospital and Harvard Medical School, Boston, Massachusetts; T. Crawford McAslan, M.D., of Baltimore City Hospi­ tals and The Johns Hopkins Medical School; lain L. Mackenzie, M.D., of York Hospital, York, Pennsylvania, and Bae­ khyo Shin, M.D., Lucille Ann Mostello, M.D.. and Martin Helrich, M.D., all of the University of Maryland Hospital and Medical School, Baltimore. Particular thanks are due Martin Helrich, M.D., Chairman, Department of Anesthesiol­ ogy for his support and encouragement.

Second Edition. We gratefully ac­ knowledge help from John New, B.A., with preparation of statistical data and we are grateful to Beverly Sopp and her staff (Lynn Kesselring and Eina Segal) for editorial assistance. Marlene Wheeler and Ruth Allan were unflagging typists. Justina Smith prepared the graphs in Ap­ pendix I and Appendix IV and assisted in ix

•

many ways in the completion of this sec­ ond edition. We are grateful to all of the reviewers of the first edition and hope to have answered their criticisms with the second edition. We are especially thank­ ful to Nancy Klemic who, while not in­ volved as an author in the second edition because of other commitments, skillfully

CONTENTS

reviewed each chapter with thoughtful criticism. In addition. George Barnas, Ph.D. provided excellent comments on Chapter 7. Finally, we thank our editor at Williams & Wilkins, Carol Eckhart, who cajoled and persuaded us sufficiently fre­ quently to get the revision finished.

Contents Preface to the Second Edition . . . . ........ . ... . . . ..... . . ..... . Preface to the First Edition . ... . .. .. ........... . .. .. . . . ...... . Acknowledgments ... .... . .. . Chapter 1

v vii

. .... , , .. , .. . . .. ... . .... ....

ix

History and Literature Review of Chest Physiotherapy, ICU Chest Physiotherapy, and Respiratory Care: Controversies and Questions ................................................ .

1

Colin F. Mackenzie. M.B., Ch.B., F.F.A.R.C.S. Chapter

2

Clinical Indications and Usage of Chest Physiotherapy: Anatomy, Physiology, Physical Examination, and Radiology of the Airways and Chest ................................................ .

53

Colin F. Mackenzie, M.B., Ch.B., F.F.A.R.C.S. Chapter

3

Postural Drainage, Positioning, and Breathing Exercises

93

Nancy Ciesla, B.S., P.T. Chapter

4

Percussion and Vibration. . ....... . ......... . . ... . ...... . ....

134

P. Cristina 1m Ie, M.S., P.T. Chapter

5

Methods of Airway Clearance: Coughing and Suctioning . . . . . . ...

153

P. Cristina Imle, M.S., P.T. Nancy Klemic, B.S.. P.T. Chapter

6

Changes with Immobility and Methods of Mobilization

188

P. Cristina Imle, M.S., P.T. Nancy Klemic, B.S.. P.T. Chapter

7

Physiological Changes Following Chest Physiotherapy. . .........

215

Colin F. Mackenzie, M.B.. Ch.B.. F.F.A.R.C.S. Chapter

B

Chest Physiotherapy for Special Patients

251

Nancy Ciesla, B.S., P.T. Chapler

9

Adjuncts to Chest Physiotherapy

281

P. Cristina 1m Ie, M.S., P.T. Chapler 10

Undesirable Effects, Precautions, and Contraindications of Chest Physiotherapy . . . ..... . .. . . . . . . . . . . . . . . . . . . . . ...... . .....

Colin F. Mackenzie, M.B., Ch.B., F.F.A.R.C.S.

xi

321

CONTENTS

xii

AppendixI

Chest Physiotherapy Statistics Showing Type and Number of Patients Treated. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

345

Appendix II

Abbreviations and Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

353

Appendix III

Summary of Chest Physiotherapy Treatment and Evaluation . . .

355

Appendix IV

Duration, Type, and Frequency of Interventions in Four Critically III Patients.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

Index. . .

. . . . . . . . . . . . . . . . . . . . . . . ... .. . .

360 367

CHAPTER 1

History and Literature Review of Chest Physiotherapy, ICU Chest Physiotherapy, and Respiratory Care: Controversies and Questions Colin F. Mackenzie, M.B., Ch.B., F.F.A.R.C.S. Historical Summary and Literature Review Update of Literature Since 1980 Pathophysiology and Risk Factors for Postoperative Respiratory Complications Therapy for Respiratory Complications Chest Physiotherapy Techniques to Change Transpulmonary Pressures Use of Bronchoscopy Summary of Literature Update Since 1980 What Is Chest Physiotherapy? What Are the Objectives of Chest Physiotherapy? Chest Physiotherapy Organization Respiratory Management Indications for Intubation, Ventilation, Weaning, and PEEP Alternative Modes of Mechanical Ventilation Misconceptions About Effects of Chest Physiotherapy Conflicting Data and Points of Contention

As an introduction to this book, Chap­ ter 1 is intended to put chest physiother­ apy into historical perspective and pro­ vide a review of the literature. A brief description of the authors' understanding of chest physiotherapy and its objectives is followed by some comments on the variations found in the literature con­ cerning indications, type, usage. and du­ ration of chest physiotherapy. A chest physiotherapy program and techniques for respiratory management and assess­ ment are discussed together with some

controversial aspects of respiratory care that relate to secretion clearance. Finally the chapter summarizes conflicting data and points of contention concerning chest physiotherapy.

HISTORICAL SUMMARY AND LITERATURE REVIEW 1901

1

Willi am Ewart described the beneficial effects of postural drainage in the treatment of bronchiectasis.

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

2 1908 1910

1915

1918 1919

1924

1933

1934

Pasteur delivered the Bradshaw Lec­ ture on massive collapse of the lung. Pasteur reported on the finding of acute lobar collapse as a complication of abdominal surgery. MacMahon described the use of breathing and physical exercises in patients with lung, diaphragm, and pleural injuries sustained in World War I. Bushnell used postural drainage for patients with pulmonary tuberculosis. MacMahon used breathing exercises for patients recovering from gunshot wounds of the chest. Featherstone described the causation of postoperative pneumonia, summa­ rized the pertinent literature since 1895, analyzed the results of 1000 consecutive medical and surgical au­ topsies, compared the incidence of pneumonia after upper and lower ab­ dominal surgery, and discussed his observations as an anesthetist on the causes of postoperative pneumonia . His pertinent findings are summarized in Table 1.1. Dr. Featherstone's masterly work is impressive because its conclusions are almost all still valid, and because of the low incidence of pneumonia, which is little changed today from 1924. Jackson and Jackson wrote on the benefits of pulmonary drainage and coughing. Winifred Linton, a physiotherapist at

1934

1938 1950

1950

19521953

the Brompton Hospital, London, En­ gland, introduced "localized breathing exercises" for the thoracic surgical pa­ tient (Gaskell and Webber, 1973). Nelson recommended bronchial drain­ age for management of bronchiectasis in children. He emphasized the use of physical and radiological examination to locate the specific position of the lung lesion and to determine patient positioning for drainage. Knies recommended bronchial drain­ age following thoracic surgery. Temple and Evans defined broncho­ pulmonary segments to identify areas of the lung needing resection. Felson and Felson used the silhouette sign to localize intrathoracic lesions radiographically. Kane described pulmonary segmental localization on posteroanterior chest x-rays. He also noted that the more ac­ curately gravity was applied to the draining bronchus, the more effective was the postural drainage.

Many reports of symptomatic and physiological benefits from breathing ex­ ercises and postural drainage appeared up to 1945 (Heckscher). However, until the 1 950s lhere was little change in the incidence of aleleclasis from that re­ ported by Pasteur and no change in lhe incidence of pneumonia reported by Featherstone (1924), despite the ad-

Table 1 .1 Summary of the Causes of Postoperative Pneumonia· 1. Postoperative pneumonia occurs with grave frequency (incidence in 1924 varied from 2.7 to 8.5%). 2. Often pneumonia is not recognized and figures that purport to give its incidence are unreliable. 3. The anesthetic agent and the method of administration, except in special cases, are seldom decisive factors. 4. Age and sex are not of importance. 5. General health and local disease of the lungs may play a considerable part. 6, At operation, every care should be taken to prevent loss of heat, of fluid, and of blood and especially exhaustion from trauma to nerve tissue and to highly vascular parts. 7. Infection of the lung is often by means of aspiration in the presence of certain other factors. 8, Severe sepsis in the other regions affects the lung via the blood stream. 9. There is evidence that lymphatic infection through the right half of the diaphragm leads first to pleurisy and then to pneumonia. 10. In the absence of severe sepsis, operations on the abdomen, and especially the upper abdomen, provide the start of the chain of events which leads to pneumonia, 11. Pain in the abdomen from operative trauma, or from inflammation, gives rise to rigidity of the anterior abdominal wall and to reflex inhibition of the diaphragm, together with some spasm of the lower intercostal muscles. The lower lobes of the lungs, then, do not freely expand and contract, so that congestion of the blood with edema sets in. 'From Featherstone (1924).

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, ANO RESPIRATORY CARE

vances in surgery, anesthesia, and anti­ biotic usage. Pioneering work on the effects of chest physiotherapy was pub­ lished by Palmer and Sellick in 1953. They described the use of breathing ex­ ercises, postural drainage with vibratory and clapping percussion, and inhalation of isoprenaline before and after surgery. This regime was significantly more effec­ tive than breathing exercises alone in re­ ducing pulmonary atelectasis in 180 pa­ tients operated upon for hernia repair or partial gastrectomy. They also found that isoprenaline combined with postural drainage and vibratory and clapping per­ cussion prevented atelectasis, but that neither intervention alone prevented it. These studies included controls but were based on the somewhat subjective data of atelectasis, as judged by clinical exami­ nation and chest x-ray. Thoren (1954), using diaphragmatic breathing and deep breathing while side-lying, postural drainage, and coughing, showed that without the use of inhalation therapy it was also possible to produce a significant reduction in pulmonary complications after cholecystectomy. Atelectasis devel­ oped in 1 1 of 101 patients treated before and after cholecystectomy, in 1 8 of 70 pa­ tients (25.7%) treated with chest physio­ therapy only after cholecystectomy, and in 68 of the 1 72 patients (35.9%) who were not given chest physiotherapy after surgery. Therefore. over 25 years ago, there ap­ peared to be a specific indication for chest physiotherapy in the prevention of pulmonary complications after surgery. Since then, there has been an explosion in the apparent indications for chest physiotherapy. It is the application of chest physiotherapy with lack of specific indications that has rightly promoted ad­ verse commentary. The original popular­ ity of chest physiotherapy arose because of the benefits produced in patients with retained lung secretions. However, to our knowledge, no one has shown that treat­ ment with chest physiotherapy has al­ tered the morbidity or mortality of pa­ tients with chronic lung disease, whereas chest physiotherapy for acute lung pa­ thology in a previously normal lung may produce a more favorable outcome. The advent of controlled mechanical

3

ventilation in 1 953 (Crampton Smith et a\., 1 954) as a means of treating acute res­ piratory insufficiency also gave rise to the realization that during artificial respira­ tion there is a special liability to pulmo­ nary complications. Opie and Spalding (1958) produced a review of some of the physiological changes that occurred dur­ ing chest physiotherapy and controlled mechanical ventilation, using intermit­ tent positive pressure with a negative ex­ piratory phase. The negative phase was then popular because this improved ve­ nous return to the heart. Advocates of in­ termittent mandatory ventilation make a similar claim about the spontaneous breath with this mode of ventilation. Opie and Spalding noted that the rate of air flow in and out of the lungs was de­ pendent upon the esophagotracheal pres­ sure gradient, and that a rise in esopha­ geal pressure with chest compression accelerated expiration (Fig. 1 . 1 ). Appli­ cation of chest compression late in expi­ ration caused only a very slight alteration in air flow. Two mechanisms for the ac­ tion of chest physiotherapy were postu­ lated. ( 1 ) By raising intrathoracic pres­ sure as a whole, air was rapidly expelled from the lungs, carrying secretions with it, as in a cough. The paper, however, ar­ gues quite successfully against this mechanism. (2) By local compression of the lung underneath the physiothera­ pist's hands, secretions were pushed from the more peripheral airways into the main bronchi. This has since been disputed as the mode of action (Laws and Mclntyre, 1 969). Wiklander and Norlin (1957) compared 1 00 patients who, following laparotomy, received chest physiotherapy, with 100 who did not. Chest physiotherapy was given before and after surgery, for usu­ ally 3 days, or as long as sputum was ob­ tained. The incidence of atelectasis was 1 3% in those who received chest phys­ iotherapy, and 24% in those who were asked to turn from side to side in bed and given instruction and help in coughing. In a frequently quoted study on the value of lung physiotherapy in treatment of acute exacerbations in chronic bron­ chitis, Anthonisen and his colleagues (1 964) compared conventional treatment and chest physiotherapy to conventional

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

4 TRACHEAL 020 '"

I

E

u

w

I sec. Figure 1 . 1 . The effect of chest physiotherapy on tracheal and esophageal pressure in a patient ventilated with intermittent positive/negative pressure ventilation. Hatched area. inspiratory period. (A) Chest vibration during inspiration. (B) Chest vibration during the inspiratory-expiratory junction. (C) Chest vibration during the middle of the expiratory period. Note the greatest trach eoesophageal pressure differences occur with Band C which are normally used methods of applying chest vi­ bration. (Tracing from Opie LH. Spalding JMK: Lancet 2.�71-674. 1958.)

treatment alone. No differences in out­ come were found between the groups randomly treated with and without chest physiotherapy. During an acute flare-up of chronic bronchitis, chest physiother­ apy did not seem beneficial. but they did not exclude the possibility of benefit in lobar atelectasis. In 1 966. Holloway and his colleagues reported that chest physiotherapy ap­ peared to cause a fall in arterial oxygen­ ation when applied to neonates with tet­ anus. This preliminary observation was followed by the publication of a study on 22 patients with tetanus who received chest physiotherapy (Holloway et a!., 1 969). These patients were compared to a matched group of 14 spontaneously breathing patients and a group of 1 5 ne­ onates receiving mechanical ventilation but not chest therapy. Chest physiother­ apy, which took the form of clapping and compression. percussion and vibration, was followed by suctioning. A fall in PaO, occurred after chest physiotherapy, but it is doubtful if the changes of PaO, 50.6 ± 6.4 to 47.0 ± 6.4 mm Hg (mean ± SO) were clinically significant, although it was apparently statistically so. The control group was ventilated but did not receive chest physiotherapy and was not turned to the same positions. Therefore. simple V /0. changes cannot be excluded

as the cause for the fall in PaO, in the treated patients. Further information about chest phys­ iotherapy was published in 1969 when Laws and Mcintyre described changes in gas exchange and cardiac output associ­ ated with chest physiotherapy in six pa­ tients in respiratory failure. All were ventilated with volume ventilators and a tidal volume (V,) of 10-13 mllkg. Cardiac output was measured with the dye dilu­ tion technique. Since this was before the era of flow-directed pulmonary artery catheters, neither pulmonary artery pres­ sure nor mixed venous gases were mon­ itored. However, mixed expired and inspired gases were analyzed. Alveolar­ to-arterial tension gradients for both 0, and CO, were derived and used to mea­ sure the efficiency of gas exchange be­ fore, during, and after chest physiother­ apy. The procedures performed included postural drainage with percussion, shak­ ing, and vibration. Artificial coughs were given in the supine position, and both lat­ eral positions, and chest compression was performed during expiration. These procedures were followed by lung hyper­ inflation (V" 20-25 mllkg). The patients were then suctioned: none had large amounts of sputum. This factor appeared to be crucial, as these authors were un­ able to show any improvement in gas ex-

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

change. They also cast doubt on the hy­ pothesis that external chest compression squeezes secretions from completely oc­ cluded airways by direct lung compres­ sion. They suggest that the amount of compression required to do this would produce areas of collapse. As an alterna­ tive hypothesis. they put forward the idea that airway clearance requires some expansion of distal lung units. Indirect ventilation of these distal airways may be achieved by collateral channels. The more proximal airways are then cleared by increased expiratory flows. generated by the physiotherapists. from the dis­ tal airways. Our explanation of the mechanism of chest physiotherapy action is similar and is described in Chapter 7. In the patients studied by Laws and McIntyre during physiotherapy. cardiac output varied up to 50% from the levels obtained before physiotherapy. These variations persisted for as long as 30 min. The greatest variation occurred during the artificial cough with inflation pres­ sures of 60-100 cm H,O. In some patients. cardiac output fell due to impaired ve­ nous return during this maneuver. In those who were conscious, the procedure was also found to be extremely unpleas­ ant. During resistance to lung hyperinfla­ tion. and with patient apprehension, car­ diac output rose. These hyperinflations. although causing such changes in the cardiovascular system, were not able to produce any lasting benefit to pulmonary gas exchange. These are some of the rea­ sons for our omission of lung hyperinfla­ tion from physiotherapy treatment (see p. 225). Lorin and Denning (1971 ) found that postural drainage produced more than twice the volume of sputum as an equal period of cough alone in 1 7 patients with cystic fibrosis. Postural drainage lasted 20 min and included positioning for the right middle lobe. lingula. and some bas­ ilar segments of the lower lobe. The pa­ tients received percussion and vibration in each position. The volume of sputum produced when compared with the vol­ ume produced by the same patient in the sitting position, coughing every 5 min for 20 min. averaged 3.4 ml and was signifi-

5

cantly greater than that of the control of 1 . 6 ml (p < 0.0001). Lord et aJ. (1972) showed anecdotal ev­ idence for radiological and arterial blood gas improvement after chest physiother­ apy in an infant and adults. Gormezano and Branthwaite (1972a) reported the ef­ fects of chest physiotherapy on 43 adults receiving intermittent positive pressure ventilation. The patients were consid­ ered in three groups: Group I included 18 patients with no cardiac disability; Group II. 1 3 patients with cardiovascular dis­ ability; and Group III, 11 patients with respiratory failure. Chest therapy in­ cluded hyperinflation to 20 cm H,O above previous ventilator settings. man­ ual chest compression. and tracheal suc­ tioning. Duration of treatment was was from 7 to 20 min. depending on whether copious secretions were mobilized. Ar­ terial blood gases were sampled before and at 5. 15. and 30 min after cessation of therapy. Patients in Groups I and III did not show any change; Group II showed a maximum fall in PaO, of 14.9 ± 4.55 mm Hg (SE) 5 min after therapy. Within 30 min this had returned to the levels ob­ tained before therapy. Hyperinflation caused a rise in PaO, in all groups. PaCO, i ncreased in all groups, but a rebreathing circuit was used during manual chest compression. The authors postulated that during chest physiotheraEY. ( 1 ) cardiac output fell; therefore. PVO, fell; and. therefore. PaO, fell; (2) there was an in­ crease in intrapulmonary shunt; and (3) there was increased oxygen consump­ tion. Because no indications for chest physiotherapy were given. it is not known whether treatmenl was per­ formed prophylactically or for a specific indication. Since the patients were turned on both left and right sides but were not apparently posturally drained with the affected lobe or segment upper­ most, it is not certain whether chest physiotherapy produced these changes or whether they were due to changes in posture. Gormezano and Branthwaite (1972b) also studied patients treated with chest physiotherapy and intermittent positive pressure breathing (IPPB). Thirty-two chronic bronchitic patients with airway

6

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

obstruction and sputum production were divided into three groups: those with re­ versible airway obstruction (Group I), those with profuse secretions (Group II), and those with respiratory failure (Group Ill). Mean PaO, fell 4-6 mm Hg in all groups, but the fall was greatest in Group II. The fall was thought to be due to in­ creased intrapulmonary shunt. However, no specific cause of the increased shunt was identified, although several were hy­ pothesized, such as decreased pulmonary artery pressure with rest after therapy, increased pulmonary artery pressure due to i ncreased cardiac output, and abolition of hypoxic pulmonary vasoconstriction. In 1 973, Clarke and his colleagues re­ ported the effects of sputum on pulmo­ nary function. Patients with copious spu­ tum production and airway obstruction (forced expired volume in 1 sec (FEV,)! forced vital capacity (FVC) < 70% pre­ dicted) improved in all measured param­ eters, particulary specific airway conduc­ tance, following sputum removal. There was, however, no relationship between the volume of sputum produced and the improvement of pulmonary function. They concluded that although sputum volume production is important, its dis­ tribution within the bronchial tree and its viscoelastic properties may be of greater importance. A report in which removal of inhaled radioactive tracers was used to measure pulmonary mucociliary clearance in cys­ tic fibrosis appeared in 1 973 (Sanchis et al.). Despite previous beliefs, mucociliary transport in 1 3 children with cystic fibro­ sis was found to take place at a similar rate to that found in normal adults. The theory that the viscid secretions found in cystic fibrosis (or mucoviscidosis) were inadequately cleared, resulting in blocked airways, stasis, and resultant in­ fection, appeared to be considerably set back by this finding (Waring, 1 973). How­ ever, one problem with the technique used was that the particle size of 3 I'm was perhaps too large and, therefore, the radioactive particles did not penetrate the lung effectively. More central pene­ tration occurred in children than in adults. Because mucociliary clearance is faster from larger airways than from smaller airways, the children may only

appear to have normal mucus clearance. The radioactive tracer clearance tech­ nique is now the accepted model for fur­ ther investigation. However, peripheral deposition of radioaerosols is difficult in patients with chronic lung diseases be­ cause they often have impaired inhala­ tion and they cough. Radioaerosols do not reach airways that are obstructed so that although airway clearance may occur, this is not demonstrated by the ra­ dioaerosol clearance technique. In 1974, the National Heart and Lung Institute, as i t was then called, organized a confer­ ence, frequently referred to as the Sugar­ loaf Conference, on the scientific basis of respiratory therapy. This conference and a similar conference on in-hospital respiratory therapy published in 1 980 are summarized in Chapter 7, pp. 245-248. Campbell and his colleagues (1975) re­ ported that bronchoconstriction, as mea­ sured by a fall in FEV" occurred in seven patients with exacerbation of chronic bronchitis following chest percussion or vibration. They found that bronchocon­ striction was particularly noticeable in patients who did not have copious spu­ tum production. The fall in FE V, was not confirmed by other studies of chest phys­ iotherapy and chronic bronchitis (Coch­ rane et al.. 1977; Newton and Stephen­ son. 1 978; May and Munt. 1979). Tecklin and Holsclaw (1975) found that following postural drainage, percussion, vibration and coughing in 26 patients with cystic fibrosis, significant increases occurred in peak expiratory flow rate . FVC, expiratory reserve volume and in­ spiratory reserve capacity. Larger air­ ways appeared to be the sites of this ben­ eficial action. There was no indication that these benefits lasted beyond 5 min after treatment had ceased. Cystic fibro­ sis is one of the few chronic lung diseases for which the benefits of chest physio­ therapy are documented. A conference in Europe, published in 1 977. summarized the state of the art (Baran and Van Bo­ gaert, 1 977). Objective evidence of change in the lungs following sputum removal by chest physiotherapy in mechanically venti­ lated patients was reported by Winning and colleagues (1975). They estimated al-

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPI RATORY CARE

veolar pressure by means of a retard mechanism applied to the lungs at end expiration. A significant fall in "alveolar pressure" was noted to occur after chest physiotherapy in 1 7 patients. Unfortu­ nately the adjustment of the ventilator necessary to produce the "alveolar pres­ sure" alters the characteristics of the lung under study. Therefore, it is difficult to determine whether the changes found were due to chest physiotherapy or ven­ tilator manipulation. The additional effect of only a muco­ lytic agent, or a bronchodilator and a mu­ colytic agent, on arterial oxygenation following chest physiotherapy was com­ pared to the therapy alone. No differ­ ences were found (Brock-Utne et al.. 1975). A similar finding was reported in which clearance of inhaled polystyrene particles tagged with technetium-99m was used to assess removal of lung secre­ tions in a double-blind crossover trial in 16 patients with chronic bronchitis (Thomson et aI., 1975). There was no sig­ nificant difference in weight or radioac­ tive content of sputum expectorated be­ tween the patients who were given S-carboxymethylcysteine, a mucolytic agent, and those who were not. Ventila­ tory capacity as assessed by dry spirom­ etry was not changed, nor was there sub­ jective improvement noted by the patients. Roper and colleagues (1976) found right upper lobe atelectasis oc­ curred after tracheal extubation in 18 of 188 newborn infants. This was thought to result from the anatomical positions of the right upper lobe bronchus and dam­ age from suction catheters. The atelecta­ sis could usually be expanded by chest physiotherapy. If it was unresponsive. an orotracheal tube was inserted, and the lungs manually hyperinnated and suc­ tioned until all secretions were mobi­ lized. Following this the trachea was immediately extubated. This regime re­ sulted in the elimination of recurrent at­ electasis as a major problem after extu­ bation. Tecklin and Holsclaw (1976) found that N-acetylcysteine (Mucomyst) and bronchial drainage and coughing produced the same changes in respira­ tory function in 20 patients with cystic fi­ brosis, that occurred without the use of the mucolytic agent. In fact, maximal

7

midexpiratory now rate worsened, show­ ing significant decrease with the inhala­ tion. This was thought to be due to renex small airway constriction and edema fol­ lowing N-acetylcysteine or due to cough­ ing. Using technetium-99m, Pavia et al. (1976) found that a mechanically vibrat­ ing pad did not significantly alter clear­ ance of sputum when compared in 1 0 patients who had histories o f productive cough and difficulty expectorating phlegm; however, postural drainage was not used. Martin et al. (1976) investigated the ability of unilateral breathing exercises to alter distribution of ventilation and blood now in patients u ndergoing bron­ chospirometry. In no instance was the distribution of ventilation or blood flow altered to the side that was supposed to be limited. However, although these pa­ tients had active tuberculosis, there was no indication that the pathological lung was the target of the therapy, since both sides of the chest were treated. All pa­ tients had less than 1 5% involvement of the lung fields by chest x-ray. respiratory function tests were all normal, and only one subject was thought to have moder­ ately advanced disease. Therefore, it is possible that in major lung pathology or in patients with chest splinting due to pain, breathing exercises may have a dif­ ferent effect, when large differences in lung/thorax compliance occur. Removal of sputum by chest physio­ therapy produced an improvement in specific airway conductance in 17 of 23 patients with chronic cough, airway ob­ struction. and at least a 30-ml sputum production per day (Cochrane et aI., 1 977). This improvement did not occur in 4 normal subjects, nor in 8 of the study patients who, on the following day, were given 1 50 mg isoprenaline base by inha­ lation instead of physiotherapy. Coch­ rane and his colleagues reiterated their belief that the distribution of sputum throughout the airways appeared to be more relevant than sputum volume, vis­ cosity. or character. No correlation was found between changes in specific air­ way conductance and sputum volume produced by chest physiotherapy. By using transcutaneous 0, monitor­ ing, the effect of chest physiotherapy on

8

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

PaO, in 45 patients who had undergone abdominal surgery was compared to three deep breaths using incentive spi­ rometry, a mechanical lung insufflator, and the blowing up of a paper coil (Hed­ strand et al" 1 978). Chest physiotherapy produced a greater increase in PaO, than did the other maneuvers, though it is doubtful if a 7 mm Hg rise in PaO, is clin­ ically any different from the 3 to 4.5 mm Hg obtained with the respiratory therapy devices. The reliability of transcutaneous 0, monitoring, when used in adults, is also in question. This paper does not re­ cord why the patients needed therapy. The respiratory therapy devices may have been used in the recommended manner; however, chest physiotherapy, which apparently consisted of ten deep breaths and a minute of coughing fol­ lowed by assisted costal breathing in the lateral position, at our institution would be considered inadequate to clear re­ tained secretions. Two abstracts (Finer et aI., 1 977; Fox et aI., 1 977) that described chest physio­ therapy for the neonate were published as papers the following year. Finer and Boyd (1978) studied 20 neonates with a mean weight of 2.07 kg. Seven neonates were mechanically ventilated; all had respiratory failure and were receiving supplemental 0,. Respiratory failure was due to respiratory distress syndrome in 14 neonates, tachypnea in 2, pneumonia in 3, and apnea in 1. Arterial blood gases were analyzed before, and 1 5 min after, postural drainage and suction (10 infants) or postural drainage, percussion and suc­ tion (10 infants). The neonates showed a rise in PaO, when postural drainage, per­ cussion and suction were used but no sig­ nificant change with postural drainage and suction alone. The same findi ngs, in a population of a different age and venti­ lated differently, were reported in the ab­ stract. It is not clear why some patients, whose data appeared in the abstract, were omitted from the paper. Fox and his colleagues (1 978) studied 1 3 newborns to "determine the benefit/ risk ratio of chest physiotherapy." All were int ubated, breathing spontaneously with positive airway pressure, and were recovering from respiratory disease (res­ piratory distress, 1 0; aspiration, 2; apnea,

1 ). After a control period, 30 sec of ante­ rior chest wall vibration was performed by using a mechanical vibrator. The in­ fants were then suctioned and hyperven­ tilated for ten breaths. Since neither pos­ tural drainage nor percussion was used. the treatment given was not strictly chest physiotherapy. However, there was a consistent trend in which compliance and functional residual capacity (FRC) increased in parallel throughout all the periods of study. Inspiratory airway re­ sistance was noted to fall significantly following chest vibration and suctioning, but this had returned to control levels within 2 hr. Arterial oxygenation fell sig­ nificantly following suctioning. This was reversed with hyperventilation. Two hours following therapy, PaO, levels did not differ from control. The fall in PaO" which was as high as 81 mm Hg in a pa­ tient breathing 55% oxygen, was not thought to be due to atelectasis because there was no change in FRC and no fall in lung compliance. It was, perhaps, due to the rise in pleural pressure accompa­ nying coughing and suctioning which may have increased a right-to-left shunt. Mackenzie et al. (1978) studied 47 pa­ tients with a variety of chest x-ray changes that included atelectasis, pneu­ monia, or lung contusion. Eight of the 47 patients were nontrauma and had multi­ ple pathology; the remainder were trauma patients. All patients were me­ chanically ventilated with positive end­ expiratory pressure (PEEP) (5-10 cm H,O). Changes in arterial oxygenation were prospectively studied before and for 2 hr after chest physiotherapy. No sig­ nificant changes in PaO, were found after chest physiotherapy. There were no dif­ ferences between patients with or with­ out trauma or between those treated with or without head-down postural drainage. The falls in arterial oxygenation reported by others to occur after chest physiother­ apy may be reversed by the use of PEEP. Unilobar lung pathology showed radio­ logical improvement in 74% (20/27) and multilobar pathology improvement in 60% (12/20). These radiological findings are similar to those obtained by fiberoptic bronchoscopy in patients resistant to rou­ tine respiratory therapy (Lindholm et al. 1 974).

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

Chest physiotherapy produced a differ­ ent effect on pulmonary function in pa­ tients with acute exacerbations of chronic bronchitis (Newton and Stephen­ son) than in patients with bronchiectasis and cystic fibrosis (Cochrane et al.). The 33 patients studied by Newton and Ste­ phenson, within 4 days of admission for acute exacerbation of chronic bronchitis, had an FEV,/FVC ratio of _ _ _ _ _

30 ••cond. -----_

Figure 1 .3. The movements of the rib cage (rc) and abdomen (diaphragm ab) are shown together with their sum, which is tidal volume. The trace shows obstructive apnea. rc and ab are first in phase for three breaths and then opposite in phase as upper airway obstruction becomes com­ plete. The sum falls to zero as apnea occurs. The 0, saturation (0, Sat) signal has a 1 O-sec delay so appears to go on falling after the patient shows arousal at the end of the period of obstructive apnea. EEG, electroencephalogram; EOG, electrooculogram; ECG, electrocardiogram. (From Jones JG: Pulmonary complications following general anesthesia. In Anaesthesia Review, ed ited by L Kaufman, 2nd Ed., Chapter 3, pp. 21 -38. Churchill Livingstone, Edinburgh, 1 984.)

expansion, and the postoperative fall in FRC and ERV may be due to respiratory muscle fatigue (Macklem, 1 981; Craig, 1 981; Jones, 1 984). Fatigue of the dia­ phragm was predicted by the fall in force generated at low-frequency compared to high-frequency stimulation (Moxham et aI., 1981 ). Respiratory muscle fatigue was relieved for long periods by a single un­ loaded breath (Lawler et aI., 1 979), and this may be the physiological basis for the beneficial effects sometime seen with intermittent mandatory ventilation (IMV). Phosphate depletion was associ­ ated with muscle weakness and hypo­ ventilation, and may particularly be a cause of respiratory complications in al­ coholics who receive dextrose intrave­ nously and gastric antacids containing aluminum or magnesium (Newman, 1 977). The mechanism determined by Aubier et al. (1985) was that hypophos­ phatemia impairs the contractile proper­ ties of the diaphragm during acute respi­ ratory failure. Enjeti et al. (1982) compared the effects of lung inflation of lobar and sublobar at-

electasis on vascular pressure-flow rela­ tionships in pigs with closed chests. With lobar atelectasis the atelectatic lobe be­ haved as a vascular resistor in parallel with the surrounding lung. When normal l ung surrounding lobar atelectasis was inflated, its vascular resistance was in­ creased and pulmonary blood flow was redistributed to the atelectatic lobe, caus­ ing a significant increase in intrapulmo­ nary shunt. However, with sublobar at­ electasis, lung inflation did not cause redistribution of blood flow. The expla­ nation of these differences was that re­ gional volume inhomogenicity occurs, which distorted the sublobar vessels and prevented redistribution of pulmonary blood flow. Enjet et al. suggested that their findings supported the use of selec­ tively applied PEEP to diseased lung units in unilateral lung or lobar disease. There was, however, no discussion of the anatomic differences between pig and human lung. Collateral airways provide a means of ventilation to atelectatic sublo­ bar units in humans but collateral air­ ways are lacking in pigs.

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

Ford e t al. (1983) studied diaphrag­ matic function in 15 patients before and after cholecystectomy. Diaphragmatic function was assessed by changes in transdiaphragmatic pressure swings dur­ ing quiel tidal breathing, the ratio of changes in gastric to esophageal pressure swings, and the ratio of changes in ab­ normal rib cage diameters. There were significant falls in transdiaphragmatic and gastric-to-esophageal pressure changes after surgery, which reverted to­ ward normal within 24 hr. Ford et al. sug­ gest this reduction in diaphragm function may be responsible for the atelectasis, re­ duced vital capacity, and hypoxemia seen in patients after surgery. Different anesthetic techniques were used but not described. No assessment of neuromus­ cular function was made to confirm that complete reversal of muscle relaxants was present. Because the role of pain in the etiology of postoperative respiratory dysfunction was still not fully established, Simon­ neau et al. (1983) examined diaphragm function after upper abdominal surgery. Opiate epidural analgesia on the first postoperative day did not modify dia­ phragmatic dysfunction, and measures of diaphragmatic function took until the seventh postoperative day to return to normal. They were able to demonstrate that the postoperative dysfunction was due to the upper abdominal surgery, not general anesthesia. Simonneau and asso­ ciates implicated neuromuscular dys­ function or impairment of diaphragmatic mechanics induced by surgery as a pos­ sible mechanism of diaphragm dysfunc­ tion. Pain relief alone did not result in re­ covery of postoperative respiratory muscle abnormalities. Schur et al. (1984) measured pulmonary function in chil­ dren before and after scoliosis surgery and compared them to a similar group of patients undergoing elective peripheral surgery. The peripheral surgery group had no postoperative change in lung vol­ umes, whereas the scoliosis patients had 44% of their preoperative VC, 81% of FRC, 124% of preoperative residual vol­ ume, and 61 % of total lung capacity. Schur et al. concluded that postoperative lung volume abnormalities were related

15

to the site and magnitude of surgery and associated abnormalities such as pain and preoperative respiratory dysfunc­ tion. The mechanism of reduced vital capac­ ity, hypoxemia and atelectasis that oc­ curs after upper abdominal surgery re­ mains unknown. However, Ford and Guenter (1984) quote Pasteur (1908, 1 910) that a defiCiency of inspiratory power is important and suggest that respiratory muscle activity may be reflexly modified by intraabdominal afferent nerves. Post­ operative diaphragm function is im­ paired after cholecystectomy and breath­ ing is predominantly thoracic. The return to normal thoracoabdominal breathing takes 24-48 hr. Animal studies suggest that when diaphragm function is im­ paired, expiratory muscles are activated and cause lung volumes to cycle below FRC. Therefore, despite loss of diaphrag­ matic contractility, there is still passive movement of the diaphragm. Expiratory muscle activation may cause atelectasis and hypoxemia because of small airway closure. Ford and Guenter (1984) suggest that the reduced tidal excursions of the lung adjacent to the diaphragm lead to the retained secretions, atelectasis, and infection seen in the lower lung fields. Intervention to block the reflex pathways responsible for decreased diaphragm ac­ tivity may be possible when the afferent limb is established. Because diaphragm activity spontaneously reverts to normal, this definitive therapy will probably only be required for 24-48 hr. Fletcher and Larsson (1985), in an in­ teresting case report, monitored expired CO, curves and sulfahexafluoride SF6 washout (used to measure FRC) in an 1 1 month-old child who developed atelec­ tasis. They found that alveolar dead space and intrapulmonary shunt were in­ creased, but the blood shunt through the atelectatic lung was too small to account for the large increase in alveolar dead space. They suggest that atelectasis in the infant is associated with a more wide­ spread disturbance in gas exchange due to the effects of interdependence on ad­ jacent lung regions. Celli et al. (1986) found that in patients with severe airway obstruction, arm ex-

16

CHEST PHYSIOTHERAPY I N THE INTENSIVE CARE UNIT

ercises increased dyspnea and lead to dyssynchronous movements of the ab­ domen and chest wall. Celli et al. and the editorial that accompanied their article, concluded that breathing and simultane­ ously exercising of the shoulder girdle muscles, which are also accessory mus­ cles of respiration, increased the de­ mands on the other respiratory muscles, resulting in an increased load on the di­ aphragm. The diaphragm works ineffi­ ciently in chronic lung diseases; hyper­ inflation and loss of elastic recoil of the lungs cause diaphragm flattening. The dyssynchronous breathing noted in the study may be due to increased dia­ phragm flattening, and diminished expi­ ratory ability of the abdominal muscles when the breathing rate increases and expiratory time decreases during ex­ ercise. Ourenil et al. (1986), after excluding neuromuscular dysfunction, concluded that postoperative diaphragm dysfunc­ tion was secondary to mechanisms such as a decrease in central nervous output. There may be inhibited phrenic nerve output due to reflexes from the chest or peritoneum. Local anesthetic nerve blockage of the sympathetic splanchnic, vagal abdominal, or afferent pathways from the esophagus or gallbladder may be therapeutic and relieve inhibition of central neural drive to the diaphragm. Craven et al. (1986) identified risk fac­ tors for pneumonia and fatality i n me­ chanically ventilated patients as univar­ iately associated with the presence of intracranial pressure monitors (p < 0.002), cimetidine treatment (p < 0.001), fall-winter hospitalization (p < 0.04), and ventilator tubing changes every 24 h r rather than 4 8 hr (p < 0.02). The overall fatality rate for 49 of 233 patients who de­ veloped pneumonia was 55%, confirming the severity of illness of these patients. The diagnostic criteria used for pneu­ monia in this study are too nonspecific and do not provide a certain diagnosis of pneumonia. However, this study shows that there is a greater incidence of a pneumonia-like disease in patients who receive steroids (as in head injury) have increased gastric pH and get more frequent ventilator tubing mani­ pulation.

The difficulties in diagnosis of nosoco­ mial pneumonia in intubated intensive care unit patients were addressed by Sal­ ata et al. (1987). They noted that patients with tracheal tubes were at particularly high risk for colonization and subsequent pneumonia because of disrupted local clearance mechanisms, underlying im­ munosuppression, frequency of invasive procedures, use of respiratory therapy equipment, and location in an intensive care environment where exposure to nu­ merous nosocomial pathogens was likely. Noninfectious pulmonary infiltrates may occur in the presence of colonization and this condition is difficult to distinguish from nosocomial pneumonia. Salata et al. studied 51 patients prospectively using three times weekly tracheal aspirates ob­ tained by a sterile suction catheter. Suc­ tion catheter sampling almost certainly increases the false positives by contami­ nation from tracheal tube to t racheobron­ chial tree, yet such sampling is fre­ quently used in the ICU. Graded gram stains, quantitative bacterial cultures, and examination of aspirates for elastin fibers were used together with clinical and radiological observation. Grading of the gram stain for neutrophils, bacteria, and intracellular organisms correlated with quantitative tracheal aspirate cul­ ture. The presence of elastin fibers pre­ ceded pulmonary infiltrates by a mean of 1.8 ± 1.3 (SO) days and had a sensitivity of 52% and predictive value of 100%. Ex­ amination of serial tracheal aspirates for elastin fibers and by graded gram stain may improve differentiation of coloniza­ tion from nosocomial pneumonia. Of the 51 patients, 34 developed pulmonary in­ filtrates on an average of 11 ± 13 days (range 1-60 days) after study entry and 21 of 34 were infected. Gram-negative ba­ cilli were the most common isolates and were found most frequently in infected patients. Faling (1988) questions how the predictive value of elastin fibers might change in patients with adult respiratory distress syndrome (AROS). Elastin fibers are produced by lung necrosis and this also occurs in AROS, making differentia­ tion from pneumonia difficult. Oriks et al. (1987) hypothesized that gram-negative nosocomial pneumonia may result from retrograde colonization

CHEST PHYSIOTHERAPY. ICU CHEST PHYSIOTHERAPY. AND RESPIRATORY CARE

of Ih e pharynx from the stomach. The in­ cidence of pneumonia in 1 32 mechani­ cally ventilated patients in an ICU was twice as high in Ihe patients given ant­ acids or histamine type 2 (H2) blockers for slress ulcer prophylaxis compared to sucralfate. Sucralfate produced signifi­ canlly lower gastric aspirate pH. Gram­ negative bacilli were isolated more fre­ quently and mortality rates were 1.6 times higher in the group of patients re­ ceiving antacids or H2 blockers. Driks et aI. suggest that sucralfate may reduce the risk of nosocomial pneumonia in me­ chanically ventilated patients because it preserves the natural gastric acid barrier againsl bacterial overgrowth and pre­ vents gastric colonization of the airway. Johanson and colleagues (1988a) exam­ ined diagnostic accuracy of bronchoal­ veolar lavage (BAL) in mechanically ven­ tilated baboons compared to culture of tracheal secretions. protected specimen brush samples. and direct lung aspirates. The cultures of the three specimens ob­ tained were compared with culture of lung homogenates. BAL produced the besl refleclion of lung infections both qualitatively and quantitatively. BAL re­ covered 74% of all microbes isolated and there were 15% false-positive specimens. Tracheal aspirates found 78% of the or­ ganisms in the lung tissue but 40% (14/ 30) of bacterial species isolated were not presenl in lung tissue. In a second paper. Johanson et aI. (1988b) compared the oc­ currence of pneumonia after topical an­ tibiotics. applied to the oropharynx and Irachea. with and without the addition of intravenous antibiotics. Thirty-five ba­ boons were studied for 7-10 days. Thirty animals had acute lung injury induced by oleic acid or hyperoxia. In 12 animals. no antibiotics or topical or intravenous antibiotics alone were given. Of the postmortem lung lobes 81 % had severe pneumonia and none was sterile. Com­ binations of topical polymyxin and intra­ venous penicillin and gentamycin were efficacious in preventing pneumonia in 23 animals. Only 15% of lobes from these animals contained moderate to severe pneumonia and 52% of the lobes were sterile. Further studies are required in humans to assess the efficacy of these re­ gimens in human subjects who may have

17

microbes with resistance to significant numbers of antibiotics. Lack of mucociliary clearance occurs in subjects with Kartageners syndrome. Vevaina et al. (1987) examined radioiso­ topic mucociliary clearance and the ul­ trastructure of respiratory cilia in such a patient. They confirmed total absence of clearance of inhaled technetium-labeled sulfur colloid after 1 hr. Transmission electron microscopy of bronchial muco­ sal biopsy tissue showed that in virtually all cilia inner dynein arms were absent. The outer and inner dynein arms on the cilia are microtubule-associated proteins that are thought to be the transducers of mechanical forces necessary for ciliary motion. Inner dynein arms are of high molecular weight and have six electro­ phoretically distinct heavy chains. These proteins have unique and high adenosine triphosphatase activity in the presence of magnesium and calcium ions. Other causes for human ciliary immotility in­ clude abnormalities of the length and basal apparatus of cilia. radial spoke de­ fects. outer or inner dynein arm defi­ ciency. absence of microtubules within the cilia. and ciliary immotility induced by infection and injury (Eliasson et al.. 1 977; Supp1 1 27. Eur / Resp Dis 1 983; Ve­ vaina et al .. 1 987). All these recognized ultrastructural defects have the same clinical effect of causing impaired ciliary clearance from the tracheobronchial tree. Tokis et al. (1987) examined whether spontaneous breathing. compared to con­ trolled mechanical ventilation. muscle paralysis. and PEEP. was an etiology i n the development of atelectasis and im­ paired gas exchange during general an­ esthesia. They used computerized axial tomographic (CT) scans and the multiple inert gas elimination technique (Wagner et al.. 1974) for assessment of the distri­ bution of V /Q ratios in 13 supine adult patients. Seven patients were smokers and two had obstructive airways disease. Ventilation of poorly perfused areas oc­ curred in nine patients ranging from 9% to 1 9% of total ventilation. CT scans after 15 min of halothane anesthesia and me­ chanical ventilation showed densities in dependent lung regions in 11 of 19 pa­ tients (Figs. 1 .4 and 1 .5). There was close correlation (r= 0.84) between the area of

18

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Awake

.7

.3

1

0 01

t

n

m

_ ZEEP

'"

01

t

n

m

""

_

PEEP p.o,

137 II'riig

/0 01

t

n

m

Figure 1 .4. Transverse computerized tomographic (CT) scans of the chest and alveolar ventila­ tion (VJ/blood flow (0) distribution (0 VA 0 liters/min), awake, during anesthesia, with con­ ventional mechanical ventilation, and after the addition of 1 0 cm H 20 PEEP. There is a unimodal V./O distribution awake with shunt (0,) of 0.8%. After induction of anesthesia, densities appear in the dependent lung and 0, is 7.4%. PEEP reduces the densities but not the shunt and causes a high V./O mode. PaD, is shown at each time. (From Tokis L, Hedenstierna G, Stranberg A, Brismar B, Lundquist H: Lung collapse and gas exchange during general anesthesia: Effects of spontane­ ous breathing, muscle paralysis and positive end-expiratory pressure. Anesthesiology 66:157 -167, 1987.) =

=

atelectasis on CT scan and the magnitude of shunt. Both CT density area and shunt increased after muscle paralysis. PEEP reduced the CT density area but did not consistently alter the shunt. During spon-

taneous breathing, shunt and density area were decreased compared to pa­ tients managed with muscle paralysis and mechanical ventilation. Tokis et al. suggested that anesthesia reduced or al-

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND R ESPI RATORY CARE

19

Awake

.9

p.o, rm'IHg

158

100 """.th.

PEEP p. o, rm'IHg

1 14

.01

t

'0.

100

Figure 1.5. Same as Figure 1 .4 except the patient is a smoker with mild airway obstruction. During anesthesia (middle panel). the diaphragm moves cranially and appears as the white area, especially in the left lung. There is, however, only a small atelectatic area and low shunt (3.4%) although there are many low VJQ areas. PEEP reduced the atelectatic area but had minimal effects on VJQ. (From Tokis L, Hedenstierna G, Stranberg A, Brismar B, Lundquist H: Lung collapse and gas ex­ change during general anesthesia: Effects of spontaneous breathing, muscle paralysis and positive end-expiratory pressure. Anesthesiology 66:1 57-167, 1 987.)

tered Ihe lone of the diaph ragm and caused development of atelectasis. The reason atelectasis increased after muscle paralysis was not explained. FRC was reduced during anesthesia, and there was a cranial shift of the dia­ phragm. During positive pressure venti-

lation, diaphragmatic movement occurs probably in a piston-like manner [K. Reh­ der, personal communication). Tokis et a!. found no further cranial movement of the diaphragm [Fig. 1 .5) after muscle pa­ ralysis, compared to spontaneous breath­ ing during anesthesia. Because the CT

20

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

scans took 5 sec, there were differences in the duration of the scan exposed dur­ ing end expiration with spontaneous breathing (rate 20/min) and mechanical ventilation (rate 1 2/min). Atelectasis is more apparent on chest x-ray at end ex­ piration, even in normal subjects. The changes in mechanically ventilated pa­ tients compared to spontaneously breath­ ing patients and unexplained differences in CT density and diaphragm position could be related to lack of standardiza­ tion of respiration and CT scans. In the experience of Tokis et aI., 73 of 78 pa­ tients studied before and after anesthesia developed atelectasis. Clearly, anesthe­ sia with halothane was associated with intraoperative atelectasis. Other inhala­ tiona I agents or anesthetic techniques such as hypnotic or narcotic agents may not produce the same incidence of atelectasis. Novak et al. (1987) reexamined the benefits of periodic hyperinflation on gas exchange for mechanically ventilated pa­ tients with hypoxemic respiratory fail­ ure. They used periodic hyperinflations of 40 cm H20 lasting 15-30 sec as a sustained, exaggerated, hyperinflation rather than a sigh. To maximize trans­ pulmonary pressure during hyperinfla­ tion, the patients were turned with the area to be expanded uppermost. Cough­ ing was encouraged during exhalation from the hyperinflation, and expiratory flow was enhanced by manual external chest compression in uncooperative pa­ tients. The procedure was performed 1 0 times a t 3D-sec intervals between hyper­ inflations, five times with the patient in each of the right and left lateral positions, and was compared to standard bag-sigh suctioning. Neither technique, alone or in sequence, resulted in changes in gas exchange or lung/thorax compliance 5 or 30 min after treatment. In view of the attempts made to in­ crease transpulmonary pressures by use of incentive spirometry and inspiratory resistive breathing devices these findings are important. The reason that 40 cm H20 was not able to recruit collapsed lung or improve oxygenation in patients with hypoxemic respiratory failure of more than 24 hr duration could be that end-ex­ piratory pressure returned to baseline

levels, and collapse reoccurred, after each maneuver. Alternatively, either al­ veolar collapse was not the cause of hyp­ oxemia, or viscid secretions blocking the airway caused overexpansion of the ven­ tilated lung that further compressed the collapsed segments. Segmental bronchi blockage could be the mechanism, as blockage at a subsegmental or more pe­ ripheral airway would allow expansion through collateral airways and a cut off in the chest x-ray air bronchogram should be clearly visible. Novak et al. do not explain why the patients were turned on both left and right sides. Respiratory mechanics and gas exchange for unilat­ eral lung lesions may improve in one lat­ eral position and deteriorate in the other. For either hyperinflation or bag-sigh suc­ tioning to be beneficial, secretions should be removed. The quantity of sputum re­ moved was not described. Mankikian et al. (1988) measured the effects of thoracic epidural block on dia­ phragm function in 13 patients after upper abdominal surgery. Fourth tho­ racic vertebral block with 0.5% plain bu­ pivacaine reversed diaphragm dysfunc­ tion that occurred consistently after upper abdominal surgery. They sug­ gested that inhibitory reflexes of phrenic motor activity arising from the abdomi­ nal wall and viscera may be involved in diaphragm dysfunction. They were un­ able to discriminate between potential inhibitory afferents from intraabdominal structure afferents and mechanical an­ tagonism between abdominal muscles and the diaphragm. Using the same methodology as in this study. Simmo­ neau et al. (1983) showed that postoper­ ative diaphragm dysfunction was un­ changed with epidural opiates. Clergue et al. (1984) showed that shallow and rapid breathing that occurs after upper abdominal surgery was not modified by spinal morphine despite complete pain relief. It appears that pain is not the mechanism underlying respiratory dys­ function after upper abdominal surgery. Epidural block with 0.5% bupivacaine is the only technique to improve respira­ tory dysfunction after upper abdominal surgery. Clearly, studies examining the effects of continuous epidural anesthesia with lower concentration of bupivacaine

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND R ESPIRATORY CARE

are warranted to determine whether Ihe incidence of posloperative respiratory complications can be reduced, Therapy for Respiratory Complications

Since 1980, the reports on therapeutic interventions that may reverse respira­ tory complications or improve prognosis and reduce hospital stay for patients with respiratory complications have magni­ fied, In this section, papers on chest phys­ iotherapy and cough, techniques to change pulmonary pressure, and use of bronchoscopy, are critically summarized, Papers published since 1980 are com­ pared to similar previous publications. Chest Physiotherapy

Weller et al. (1980) found that chest physiotherapy resulted in significantly improved peak now rates in 20 children with cystic fibrosis. No patients received aerosol therapy. On average, about 8 ml of sputum was produced. In 1 2 patients sputum grew Pseudomonas aeruginosa. Pulmonary function was assessed for 24 hr and compared to no chest physiother­ apy. The authors suggest that central but not peripheral airway clearance occurred with chest physiotherapy. Bronchodila­ tors had no added effects. Tecklin and Holsclaw (1975) also reported increased peak now rates but did not follow up their 26 patients with cystic fibrosis and had no control group. In a study designed to eliminate dis­ crepancies in arterial blood gases due to changes in position. Holody and Gold­ berg (1981 ) examined the effect of me­ chanical vibrations over the anatomic area of acute lung disease during therapy for atelectasis or pneumonia. Patients were seated upright or in a high semi­ Fowler position. Ten patients were stud­ ied, at least nine of whom had lower lobe lung pathology. Mechanical vibration lasted 30 min and the patients were suc­ tioned only after completion of chest vi­ bration therapy. Blood gases at 30 min and 1 hr after therapy showed average in­ creases (p < 0.05) in PaO, of 10 and 1 5 mm Hg, respectively. when compared to baseline. In these acutely ill patients, 70% of whom were mechanically venti-

21

lated, the authors suggested that vibra­ tion therapy may be applied when pa­ tients were intolerant of postural drainage and percussion. There was no indication of how much sputum was pro­ duced with suctioning, no details of use of PEEP, and no information on the du­ ration of improved PaO, beyond 1 hr. If the data from the three spontaneously breathing patients are extracted, there are probably no significant differences in PaO,. No difference was found between chest physiotherapy and IPPB in prevent­ ing postoperative pulmonary complica­ tions among patients who underwent upper abdominal surgery (Schuppisser et aI., 1 980). Because of the potential haz­ ards with IPBB of cross-infection. gastric dilatation, ileus, tension pneumothorax, decrease in FRC and PaO" and increased airway resistance, hypotension, and gas­ trointestinal perforation, the authors rec­ ommended chest physiotherapy (see Chapter 9). Kigin (198 1 ) published a com­ prehensive review of chest physical ther­ apy for the postoperative or traumatic in­ jury patient. She suggests that controlled studies of secretion removal techniques are a priority for this group of patients. She identifies the lack of studies deter­ mining the beneficial components of chest physiotherapy and states that clar­ ification of contraindications and compli­ cations is required. Kerrebijn et al. (1982) were unable to show clearance of mucus from peripheral airways in 25 clinically quiescent, spu­ tum-producing children with cystic fi­ brosis. Chest physiotherapy or no ther­ apy with or without aerosolized N­ acetylcysteine was randomly compared on consecutive days. No effects on respi­ ratory nows (maxium expiratory, FEV" or VC) and volumes (total lung capacity) were found. One of the points discussed by these authors was that the optimum frequency and strength of chest percus­ sion were not the same at different ages because of changes in chest wall compli­ ance and amounts of lung tissue. Patients with cystic fibrosis may respond differ­ ently to chest physiotherapy when in a quiescent, compared to an active phase. The papers in which beneficial effects oc­ curred in patients with cystic fibrosis

22

CHEST PHYSIOTHERAPY I N THE INTENSIVE CARE UNIT

(Tecklin and Holsclaw; Weller et al.) do not identify whether their patients were experiencing exacerbations at the time of therapy. The differences in techniques (frequency of percussion and force and frequency of chest vibration) may also in­ fluence the response of individual pa­ tients to chest physiotherapy. These techniques are discussed in Chapter 4. The conclusion of Kerrebijn et al. that no clearance of mucus from peripheral air­ ways occurred in patients with cystic fi­ brosis confirmed the findings of Weller et al. Rossman and colleagues (1982) com­ pared the effectiveness of spontaneous cough (control). postural drainage (PD) with and without mechanical percus­ sion. PD deep breathing with vibration and percussion administered by a phys­ iotherapist. and directed vigorous cough. They found PD was not as effective as physiotherapy or cough and frequent self-directed vigorous cough was most ef­ fective. This was in contrast to the find­ ings of Sutton et al. (1983) who report that the addition of PD nearly tripled the spu­ tum yield achieved by cough alone. Ross­ man et al. did not differentiate between the use of a forced expiratory technique (FET) and cough. Sutton et al. (1983) found that cough alone did not enhance mucus clearance. but FET and PD did. The forced expiratory technique con­ sisted of one or two forced expirations from mid- to low-lung volume. followed by a period of relaxation and diaphrag­ matic breathing. No glottic closure oc­ curred and the airway compressive phase that characterizes coughing was avoided so that worsening of broncho­ spasm may be prevented by this forced expiratory maneuver (Sutton et al.. 1 983). (See Chapter 3, p. 1 2 1 . ) In 1 0 patients with copious sputum (mean 63 ml/24 hr) FET alone and with PD i ncreased the clearance of sputum compared to control. Directed coughing without FET and PD was not different from control. Sutton et al. (1984) suggest that protracted cough­ ing was fruitless and exhausting. leading to poor compliance. Newhouse (1984) argued that the dif­ ferences between the findings of the two studies investigating clearance of mucus probably related to secretion volume and

rheology. Newhouse suggested that cough is more effective in clearing bo­ luses of secretions rather than thin, wa­ tery. copious secretions. which were more likely to be cleared by PD. The con­ flicting results of Oldenberg et al. (1979). Bateman et al. (1981 ), and Rossman et al. (1982) concerning the role of cough in­ dependent of the other maneuvers used in chest physiotherapy stimulated De­ Boeck and Zinman (1984) to assess con­ ventional pulmonary function tests after cough alone or chest physiot herapy for patients with cystic fibrosis. Chest phys­ iotherapy was standardized, consisting of 2 min of percussion and vibration in 1 1 postural drainage positions that pro­ ceeded sequentially from lower to upper lobes. The patient then performed three slow vital capacity maneuvers, during which exhalation was assisted by the physiotherapist. This was compared with a directed vigorous cough session re­ peated 1 1 times over 1 0 min in the seated position. The amount of sputum pro­ duced and the pattern and magnitude of changes in PFTs 1 hr after treatment on consecutive mornings were the same after cough or chest physiotherapy. De­ Boeck and Zinman confirmed the lack of correlation between sputum produced and improvement in PFT. They sug­ gested that vigorous coughing and FET should be further investigated for long­ term effectiveness in cystic fibrosis. The studies on patients with acute and chronic lung diseases comparing the effectiveness of the independent compo­ nents of chest physiotherapy are confus­ ing, because combinations of compo­ nents may be more effective than a single variable under study. The multitude of possible permutations of chest vibration. percussion, cough. FET. PD, or tracheal suctioning in patients with different chronic or acute lung pathology makes clinically relevant studies difficult to find. The clinician wishes to know if lung reexpansion is achieved more efficiently when all the techniques are used or whether one or two of them may be omit­ ted without loss of effectiveness. To date, this dilemma is unresolved. Zapletal et al. (1983) found that chest physiotherapy that included manual chest percussion and vibration in various

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

postural drainage positions and stimula­ tion of cough produced adverse effects in 24 patients with cystic fibrosis, Despite production of 2-10 ml of sputum, flow at 25% vital capacity decreased signifi­ cantly. The authors suggest that the flow limitation may result from adverse ef-

23

fects of prolonged coughing. Narrowing of bronchi was demonstrated on contrast cinebronchiographic studies (Fig. 1.6A and B). For postoperative patients, Morran et a!. (1983) showed in a prospective ran­ domized controlled trial of 102 patients

Figure 1,6. (A) Contrast bronchograms ot the right lung of a 1 7-year-old female with cystic fibro­ sis. Arrow denotes significant narrowing of the right medial basal bronchus. On left is anteropos­ terior view and on right a lateral view during normal breathing. (6) Same as A during coughing. Note that there is marked narrowing of the airways during coughing and a lack of evacuation of the contrast material from peripheral airways. (From Zapletal A, Stefanova J, Horak J, Vavrova V, Somanek M: Chest physiotherapy and airway obstruction in patients with cystic fibrosis-a neg­ ative report. Eur JRespir Dis64:426-433, 1983.)

24

CH EST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

that routine prophylactic chest physio­ therapy significantly reduced the fre­ quency of chest infection after elective cholecystectomy. The authors differenti­ ated between routine and intensive chest physiotherapy. Postural drainage and bronchodilator therapy was referred to as intensive chest physiotherapy and breathing exercises, assisted coughing, and vibration of the chest wall were used as routine prophylactic chest physiother­ apy. The control group in their study re­ ceived only encouragement to breathe deeply and cough from nursing and med­ ical staff. The authors specifically distin­ guished bet ween atelectasis and infec­ tion. Although 18 patients receiving routine chest physiotherapy developed atelectasis, only 7 developed chest infec­ tion, whereas of patients not receiving chest physiotherapy, 1 1 had atelectasis and 1 9 developed chest infection (p < 0.02). There were weaknesses in this clinical study because previous respira­ tory disease was not adequately defined and the means for confirming infection and use of antibiotics were vague. The authors suggested that routine chest physiotherapy prevented progression of atelectasis to chest infection. Flower et aJ. (1979) showed that maxi­ mum peak vibration of the bronchial tree in adults occurred at about 1 6 Hz using an external chest pressure of about 2.5 kg. Hand clapping reaches only about 8 Hz. King et aJ. (1983), however, found that external chest wall compression at frequencies above 3 Hz and up to 17 Hz increased tracheal mucus clearance. Peak enhancement of clearance reached 340% of control at 1 3 Hz. King et aJ. sug­ gested that the high-frequency chest wall compression may cause a reduction in cross-linking of mucus glycoproteins, re­ sulting in improved mucus clearance. They did not consider other factors be­ sides frequency when comparing manual and mechanical techniques of chest physiotherapy. The clinical advantages of manual percussion over mechanical devices are in our opinion, significant, and are fully discussed in Chapter 4. Kigin (1984) comprehensively re­ viewed indications for chest physiother­ apy and the effects of the individual com­ ponents such as positioning, suctioning,

manual vibration, percussion, shaking, and cough. The use of chest physiother­ apy in acute and chronically ill patients was described. Carswell et aJ. (1 984) ex­ amined deoxyribonucleic acid (DNA) output in the sputum of patients with cystic fibrosis to determine if it was a useful indicator of pulmonary cellular damage. Aggressive therapeutic inter­ vention may be more efficiently timed to coincide with increased rate of lung dam­ age if DNA was such an indicator. How­ ever, sputum DNA showed similar vari­ ations to sputum weight and was not well correlated with lung damage. Sputum was obtained by physiotherapy using percussion and postural drainage, breathing exercises and chest shaking, assisted panting, huffing, and coughing. Physiotherapy sessions lasted 25 min. Pa­ tients carried out this therapy at home two times per day for 3 weeks, followed by 3 weeks parental use of a mechanical percussor. For the next 3 weeks a profes­ sional physiotherapist visited the home 5 days a week and provided a third session in addition to parental chest physiother­ apy. In two of the three subjects in which data were gathered, the professional and parental regimen produced significantly greater quantities of DNA than the per­ cussor or parental therapy alone. Sputum weight correlated well (p 0.90) with DNA (Fig. 1 .7). This suggests that the pro­ fessional physical therapists treatment was more effective than the parental or percussor therapy in removing secre­ tions. Buscaglia and SI. Marie (1983) found no dangerous hypoxemia using continuous oximetry monitoring of arterial 0, satu­ ration during chest percussion and vibra­ tion with postural drainage. Ten sponta­ neously breathing patients with acute exacerbation of severe COPD were stud­ ied. Although most patients had elevated PCO, and required 0, to maintain PaO, above 60 mm Hg, the largest decrease in saturation was 2%, and the lowest abso­ lute value was 91%. These findings are surprising, considering the variability of FiO, in spontaneously breathing patients receiving Oz' In addition, even in normal individuals, arterial saturation fre­ quently falls below 91% during coughing and breathholding. =

CHEST PHYSIOTHERAPY. ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

25

Eb:tstgtbc[ag:iB�gj[!]es + Percussor

Parental

+

Parental

Professional °

100

�\

(mgl24 hours)

:

80

/

60

°

°

40 :

°

•

!..

l!

°

40

SO

.

• Sputum Weight (g/24 hrs) o Sputum DNA Output

,

.

°

.

30

.

20

•

� H , t' ' vvJ \Nw

20

0

10

20

30

•

40 DAYS SO

: . '\.�\0\ .�\�N ". .,

' �. : :

60

.

10

:� :

70

Figure 1.7. Weight of sputum (solid circles) and DNA (open circles) is compared (p 0.9) in the course of parental chest physiotherapy twice a day (parental). mechanical percussor twice a day (percussor). and parental and physiotherapist treatment for a total of three treatments per day (professional) Sputum production was greater but peak ex piratory flow less with the profe ss iona l than percussor regimen. (From Carswe ll F. Robi nson OW. Ward CCl, Waterfield M R : Deoxyribo­ nucleic acid output in the sputum from cystic fibrosis palients. Eur J Respir Dis 65:53-57, 1984.) =

.

Sutton et aJ. (1985) examined the value of percussion. vibratory shaking, and breathing exercises with and without postural drainage and FET in eight pa­ tients with copious sputum production. Using inhaled aerosols, bronchial secre­ tions were labeled and their clearance monitored by gamma camera. Random­ ized treatments included vibratory shak­ ing during relaxed expiration followed by maximum inspiration. percussion during tidal breathing compared to a con­ trol period of 30 min postural drainage, and voluntary cough. Vibration (12-16 Hz), shaking (2 Hz). and percussion (ap­ proximately 5 Hz) were used to enhance mucus clearance. No differences in clear­ ance of radioisotope were found between any of the treatments. although the wet weight of sputum was increased by vi­ bration and percussion with deep breath­ ing. Why radioisotope clearance was no different when sputum clearance was in­ creased with vibration and percussion was not explained. but may be due to

particle penetration of the peripheral air­ ways. The techniques of labeling and scanning of the three regions are open to criticism because of the inability to ade­ quately visualize the isotope in three di­ mensions. The requirement that the pa­ tients are in exactly the same position with the same lung volume to make valid comparisons before and after each of the therapies that were as many as 1-4 days apart is not addressed. Although the eight patients did have copious sputum, the range varied by over 100%. The eight patients had three different causative pa­ thologies, which suggests that they might have responded differently to the thera­ pies under study. Some of the variations found in studies using radioisotopes may be due to different techniques for aero­ sol deposition and measurement of clearance. Chest percussion was found to be of lit­ tle value as an adjunct to postural drain­ age and instructed coughing in the treat­ ment of 10 patients with chronic

26

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

bronchitis (Wollmer et aI., 1 985). Percus­ sion was associated with a small decrease in FEV, after therapy; however, this was not associated with any changes in 0, saturation. Radioisotope penetration was higher following chest physiotherapy with percussion. The two patients who produced 1 00 ml and 1 30 ml of sputum had substantially higher clearance of in­ haled radioisotope particles when per­ cussion was included, compared to when it was omitted. The findings were not, therefore, clear-cut. Mazzocco et al. (1985) found that chest percussion and postural drainage were helpful, safe, and effective in assisting 13 patients with sta­ ble bronchiectasis to clear secretions. There were no adverse effects on heart rate or rhythm, pulmonary function or 0, saturation. Mackenzie and Shin (1985) measured cardiorespiratory function before, imme­ diately after, and 2 hr after chest physio­ therapy in 1 9 patients with posttraumatic respiratory failure. Cardiac index was unchanged and intrapulmonary shunt fell, followed at 2 hr by an increase in lung/thorax compliance. The reduced cardiac output after chest physiotherapy reported by Laws and Mcintyre (1969) was not found, but cardiac output was not measured during chest physiother­ apy. In addition, the patient population was young and did not have preexisting cardiac disease. None of the detrimental cardiopulmonary changes associated with bronchoscopy occurred in these critically il l patients, yet beneficial ef­ fects on gas exchange and lung mechan­ ics were documented. Klein et al. (1988) found that heart rate, systolic and mean blood pressure, and cardiac output were increased during chest physiotherapy. Metabolic rate de­ termined by measurement of 0, con­ sumption and CO, production was in­ creased. In no patient was PaO, changed but in some groups PaCO, i ncreased with chest physiotherapy. The increases in blood pressure and heart rate were atten­ uated by continuous infusion of 3 IIg/kg fentanyl but not 1.5 IIg/kg fentanyl. Fen­ tanyl did not alter the increased meta­ bolic response with chest physiotherapy. Cardiac output increased up to 50%

above baseline values in two patients who did not receive fentanyl only pla­ cebo infusion. Cardiac output was mea­ sured in 6/10 patients in the groups that received 3.0 and 1 .5 IIg/kg fentanyl. Nei­ ther dose prevented elevation of baseline cardiac output by about 20-25% during chest physiotherapy. In all instances within 1 5 min of the finish of chest phys­ iotherapy cardiac output was not differ­ ent to baseline whether the patients received fentanyl or placebo. The un­ changed cardiac output and PaO, imme­ diately after therapy confirms the find­ ings of Mackenzie et al. (1987b) and Mackenzie and Shin (1985). The issue is whether elevation of cardiac output and metabolic rate during chest physiother­ apy is clinically significant. Is this a rea­ son to withhold chest physiotherapy in critically ill patients with cardiovascular dysfunction? These are the very patients who can least tolerate pulmonary deteri­ oration. In our opinion chest physiother­ apy should not be withheld because car­ diac output and metabolic rate increase. Instead sedation and analgesia are used before and during therapy. Van der Schans et al. (1 986) examined percussion in nine patients with stable chronic airflow obstruction and exces­ sive tracheobronchial secretions. Manual percussion was no different when ap­ plied in combination with PO coughing and breathing exercises for 20 min. com­ pared to the same regimen without per­ cussion. The addition of PO and coughing with or without percussion improved mucociliary clearance. The authors sug­ gested that because manual percussion as a single procedure was found to im­ prove tracheobronchial clearance. it may be useful when a patient was not able to cough or cannot tolerate postural drainage. An editorial on management of acute bronchiolitis in infancy (Milner and Mur­ ray, 1 988) suggests that chest physiother­ apy, although a popular therapy, may not be beneficial. In a controlled study of 96 children with bronchiolitis chest phys­ iotherapy was not found to be beneficial (Webb et aI., 1 985). The criteria for as­ sessment of benefit were clinical and, therefore, subjective. The authors do not

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPI RATORY CARE

recommend chesl physiotherapy in the management of acute bronchiolitis as the babies require considerable disturbance and many children were noted to become more distressed during and immediately after chest physiotherapy. Directed, su­ pervised coughing was found to be equally effective as chest physiotherapy in management of 39 patients hositalized for treatment of an exacerbation of pul­ monary symptoms from cystic fibrosis (Bain et aI., 1 988). Directed coughing con­ sisted of five quick small huffs or pants followed by expiratory huffs and cough­ ing until all loosened sputum was cleared from the airways. The sequence was then repeated twice every 5 min for 35 min. No differences were found in pulmonary function tests and sputum charac­ teristics. Cardiac output, oxygen consumption, and arterial and mixed venous saturation (5VO,) were measured during suctioning in 10 acutely ill mechanically ventilated patients (Walsh et aI., 1989). 5uctioning produced significant decreases in 5V O, predominantly due to increased 0, con­ sumption but also due to an inadequate rise or even a fall in cardiac output. Ar­ terial saturation was virtually unchanged partly because the patients were preox­ ygenated with 100% 0, before suction­ ing. The authors comment that arterial saturation was not a sensitive indicator of changes in 5VO,. The mechanism pos­ tulated for fall in 5V O, was decreased cardiac output occurring from decreased intrathoracic pressure with suctioning. If this is indeed the mechanism, closed sheath suction catheters which prevent entrainment of atmospheric air during suctioning may potentiate this effect. The solution is to sedate agitated patients be­ fore suctioning so that 0, consumption is reduced and to restrict suctioning if no secretions are obtained. Techniques to Change Transpulmonary Pressures

By producing either subatmospheric pleural pressure or positive airway pres­ sure, bilateral lung expansion is achieved because both maneuvers in­ crease transpulmonary pressure. Me-

27

chanica I means that generate a subat­ mospheric pleural pressure greater than the normal 5-10 cm H,O include inspi­ ratory incentive spirometry, inspiratory resistive breathing, and, in patients with diaphragmatic paralysis, electrical pac­ ing. Positive airway pressure is obtained by mechanical ventilation with or with­ out PEEP, continuous positive airway pressure (CPAP) during spontaneous res­ piration, or IPPB. Pressure support and changing inspiratory/expiratory ratios may alter the length of time during the respiratory cycle that transpulmonary pressure is increased. Inspiratory resis­ tive breathing may also increase respira­ tory muscle endurance and strength. Me­ chanical means of altering local lung expansion include selective positive pressure inflation of an atelectatic lung segment, recruitment by collateral ven­ tilation, and interdependence. Anderson et al. (1979) showed that at­ electatic lung can be recruited through collateral airways by positive pressure ventilatory techniques. They suggested that time constants (product of compli­ ance and resistance) for collateral air­ ways during exhalation were longer than during inspiration, so more air entered the lung at the periphery of an atelectatic area than left during exhalation. As a re­ sult, pressure rose distal to the obstruc­ tion and became greater than the pres­ sure in the surrounding lung. Collateral airway reinflation, therefore, potentially forced secretions centrally to larger bron­ chi where they may be more easily re­ moved. Andersen et al. (1980) suggested the use of periodic applications of contin­ uous positive airway pressure (CPAP) by face mask to increase collateral ventila­ tion of obstructed lung regions. Twenty­ four patients with postoperative atelec­ tasis were studied in a prospective randomized controlled clinical trial. CPAP with 1 5 cm H,O once per hour and conventional therapy was compared to conventional therapy alone. Conven­ tional therapy included humidification, oxygen, intensive chest physical therapy with three times daily postural drainage, tracheal suctioning, and instruction in deep breathing. The comparison was made over 24 hr and successful therapy

28

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

was considered to be an increase in PaO, of at least 1 5% or decrease in area of ra­ diological atelectasis by 50%. Signifi­ cantly more patients were treated suc­ cessfully with the addition of CPAP. Chest wall strapping, such as occurs with dressings around chest tubes or after thoracotomy, reduces respiratory excursion and the ability to generate subatmospheric intrapleural pressures and increases the elastic work of breath­ ing. After strapping the chest of four healthy young men, OeTroyer (1980) found that both FRC and chest wall com­ pliance were decreased. FRC decreased about 1 liter and compliance of the chest wall fell from 22.9 to 1 0.6 mljcm H,O (Fig. 1.8). Whenever possible, chest wall restriction should be minimized as it re­ duces subatmospheric pleural pressure causing lung volume to fall below FRC. Belman and Mittman, in an editorial ( 198 1 ), raised the issue of the unproven efficacy and widespread use of incentive spirometry. They found the evidence against the efficacy of incentive spirom­ etry convincing, yet they concede that the theoretical basis for expecting this approach to be effective was also con­ vincing. Lung hyperinflation by what­ ever means reverses low lung compliFigure 1.8. Effects of chest strap­ ping on the pressure-volume curve of the relaxed chest wall in four subjects. Control data are shown as solid circles data after strapping as open circles. Transthoracic pressure shown on the abscissa is the difference between esophageal and barometric pressures. (From DeTroyer A: Mechanics of the chest wall during restrictive tho­ racic strapping. Respiration

39:241 -250, 1 980.)

ance caused by quiet breathing. They suggest that when multiple measures in addition to lung hyperinflation were used, encouraging results were obtained. Lung hyperinflation affects the high com­ pliance lung units and needs to be sup­ plemented with other measures that di­ rect ventilation to the low compliance compartments in the lung, particularly in the high-risk patients with obesity or ob­ structive lung disease. Falk et al. (1984) suggest that periodic application of face mask positive expira­ tory pressure (PEP) with forced expira­ tory technique (FET) improved sputum production and was better tolerated by 74 patients with cystic fibrosis than pos­ tural drainage, percussion and chest vi­ bration, or PO and periodic application of PEP or FET. They thought that PEP may be beneficial to other patients with ex­ cessive secretions. Celli et al. (1984) showed that lPPB, in­ centive spirometry (IS), and deep breath­ ing exercises (OBE) were superior to no treatment at all in preventing pulmonary complications in a controlled trial among 1 72 patients after abdominal surgery. Hospital stay for 21 of 81 patients who underwent upper abdominal surgery and received incentive spirometry was

': ]

2

6

�

�:> g

4

2 -10

0

10

-20 -10

0

10 20 -10 0 !'6-p. , em 11:10

-20

20

-20 -10

,

,

,

0

10

20

10

20

30

� 1O

:> -'

9 6 4 2 i

i

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

shorter than the control group who re­ ceived no therapy. The authors acknowl­ edge that the optimal treatment schedule for these techniques remains unknown and they recommend further study of IS and DBE in patients at high risk of pul­ monary complications. Many of the ar­ guments they use to support DBE would equally apply to chest physiotherapy. This was not considered. An editorial (Ford and Guenter) referring to Celli's paper stressed that the most important factor predicting the development of postoperative pulmonary complications was the upper abdominal site of surgery. It may also be important to know which organs were involved in the upper ab­ dominal surgery. Dull and Dull (1983) found that 49 adult patients who underwent cardiopul­ monary bypass and were assigned to ei­ ther mobilization, incentive spirometry, or maximal inspiratory breathing exer­ cises all experienced decreases in lung volume. No advantage was shown with incentive spirometry or breathing exer­ cises in addition to mobilization, com­ pared to mobilization alone. Stock et al. (1985) demonstrated that the use of in­ centive spirometry with documented maximal volume inhalations was not su­ perior to deep breathing exercises in pre­ venting postoperative pulmonary com­ plications after upper abdominal surgery. Respiratory function tests including FRC, FeV" and FVC were not different. CPAP produced a more rapid increase in FRC than incentive spirometry or deep breathing. The effect of continuous positive air­ way pressure (CPAP) and blow bOllles on FRC was examined by Heitz et al. (1985) in two groups of comparable patients un­ dergoing elective abdominal su rgery. Both blow bOllles and CPAP increased FRC by 50% both preoperatively and postoperatively. The significant reduc­ tion in FRC seen postoperatively re­ turned within 10 min of stopping either therapy. CPAP had a lower resistive work of breathing and was beller toler­ ated by patients. The average postopera­ tive reduction in FRC in the 20 patients who underwent upper abdominal sur­ gery was 20-30%. Both CPAP and blow bollles increased FRC, but most authori-

29

ties suggest CPAP is superior to achieve the increase. The rapid reversal of bene­ fit on cessation suggests that CPAP should be used continuously for the ther­ apy to alter outcome and reduce postop­ erative complications. Incentive spirometry was not found to be beneficial in producing differences in chest x-ray, PaO" spirometric evalua­ tion, or clinical evaluation 2 and 4 days after elective cholecystectomy, com­ pared to a group of similar patients who had no specialized respiratory care (Schweiger et aI., 1 986). Both groups of 20 patients had 1 2-16% pulmonary compli­ cations. The low-risk patients with sub­ costal cholecystectomy incisions did not benefit from incentive spirometry. IPPB is reported to improve lung compliance in subjects with kyphoscoliosis (Sinha and Begofsky, 1 972). IPPB was examined to see if it conferred benefit on patients with respiratory muscle weakness. How­ ever, in 14 subjects with either quadri­ plegia or muscular dystrophy, no imme­ diate improvements were derived in ventilatory mechanics (McCool et al.. 1 986). The study by Zibrak et al. ( 1 986) strongly suggested that a concerted effort to set priorities for the use of respiratory therapy techniques (bronchodilator aer­ osols, ultrasonic mist aerosols, IPPB, in­ centive spirometry, and oxygen therapy) for patients in at least seven common di­ agnostic categories can be successful . Im­ plementation of these priorities did not change overall mortality, although pa­ tients stayed a shorter time in the hospi­ tal after coronary artery bypass surgery and the staffing of respiratory therapists was reduced. In an editorial commenting on rational respiratory therapy stimu­ lated by the study of Zibrak et aI., Pelly (1986) identified the core of a major prob­ lem. He states that the prescribing phy­ sician's lack of firm grounding in the ra­ tional ordering of respiratory therapy is one of the factors that created the uncon­ trolled and excessive use of respiratory therapy. A large difference in usage of IS and chest physiotherapy after coronary artery surgery exists in the United States and Great Britain. Only 44% of 39 hospi­ tals used IS in Britain, whereas O'Dono­ hue ( 1 985) reported 95% used IS in the

30

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

United States. Chest physiotherapy was only used in 41% of U.S. hospitals after cardiac surgery, whereas in Britian all hospitals surveyed used CPT Uenkins and Soutar, 1 986). Strohl et al. (1984) reported on the ef­ fects of bilateral phrenic nerve pacing in a C, quadriplegic. The results are sum­ marized in Figure 1 .9. Both upper and lower rib cage showed paradoxical mo­ tion during paced breaths in the supine and 85' upright position. Abdominal compression decreased supine tidal vol­ ume 1 0-20%, whereas in the upright pos­ ture it i ncreased tidal volume 200%. Ab­ dominal compression changed only the movement of the lower ribs. These ob­ servations suggested that the diaphragm can move the lower ribs independently of the upper ribs. Tidal volume during diaphragmatic pacing is determined both by the resting length and diaphragm load. Lisboa et al. (1986) showed that unilateral diaphragm paralysis resulted in use of in­ tercostal and accessory inspiratory mus­ cle or compensatory use of abdominal expiratory muscles. Vital capacity was reduced in all cases of unilateral dia­ phragmatic paralysis. There is a decrease in inspiratory muscle efficiency that may contribute to regional V /Q mismatch, di­ minished lung volumes, and decreased endurance during inspiratory resistive loading at lung volumes higher than FRC (such as occur in emphysema). The mechanisms for the increase in 0, cost of breathing that results include changes in mechanical coupling, the pattern of reFigure 1.9. Upper and lower rib cage motion and tidal volume are shown during a postural shift from full recumbency to 85° upright pos­ ture and subsequent application of abdominal compression. The dia­ phragm is paced by phrenic nerve stimulation. (From Strohl KP, Mead JM, Banzett RB, Lehr J, Loring SH, O'Cain CF: Effect of posture on upper and lower rib cage motion and tidal volume during diaphragm pacing. Am Rev Respir Dis

130.�20-321, 1984.)

cruitment of respiratory muscles, or the intrinsic properties of the inspiratory muscles at shorter length (Collett and Engel, 1986). High-frequency ventilation produces positive airway pressure and changes transpulmonary pressures like PEEP. George et al. (1985) found that high-fre­ quency oscillation (HFO) at 8-12 Hz by means of a bass loudspeaker applied through a mouthpiece in seven normal subjects increased clearance of radioiso­ tope compared to a control group with no HFO. HFO may alter the viscoelastic properties of the mucus either by in­ creased vagal discharge or an effect on the cross-linking altering mucociliary coupling and viscoelasticity. Further studies are clearly required to confirm these findi ngs and examine the effects on patients with excessive sputum produc­ tion and purulent sputum because there is conflicting evidence of the benefits of HFO on mucus clearance. Purulent spu­ tum contains copious white blood cells, bacteria, and their products. Neutrophil elastase and bacterial products damage human ciliated epithelium and reduce ciliary beat frequency in vilro (Wilson el aI., 1 985, 1 986). Ricksten et al. (1986) st udied 43 con­ secutively randomized patients who had elective upper abdominal surgery. They compared CPAP, PEP, and a control group using a deep breathing device (Tri­ flo). One of the three therapies was ad­ ministered for 30 consecutive breaths every waking hour for 3 days postopera-

I'tf'IENIC PACED BREATHS

we !

a: :!!

"0 .. ..

"' 0:

SlftIE

-----.. , [ABDOMINAL COMPRESSION UPRIGHT------_

45 �

I

5"

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

tively. Postoperative pain was controlled using lumbar epidural morphine for at least 2 days postoperatively. Pulmonary function tests and chest x-ray were mea­ sured regularly before and after surgery. Alveolar arterial PO, difference in­ creased in all groups but was signifi­ cantly lower in the PEP group after 2 days and after 3 days with CPAP and PEP compared to control. Peak expiratory flow was not different, but FRC was higher in the PEP and CPAP groups by the third day after su rgery. Atelectasis occurred in 6 of 15 control patients, no PEP, and 1 of 1 3 CPAP palients. Ricksten et al. concluded that the simple PEP mask was equally effeclive as the CPAP system in preservalion of lung volumes and prevention of atelectasis after upper abdominal surgery. Hofmeyer et al. (1986) compared three treatment regimens for assisting clear­ ance of secretions in palients with cystic fibrosis to determine if PEP with or with­ out postural drainage increased sputum yield. Postural drainage without PEP pro­ duced more sputum than PEP and pos­ tural drainage. Both the postural drain­ age regimens produced more sputum

31

than PEP therapy in the silting position. Sputum clearance was less effective when PEP was included in the treatment regimen. Albert et al. (1987) examined the ef­ fects of the prone position in oleic acid­ induced acute lung injury in dogs. The prone position produced an immediate and persistent increase in PaD, and de­ crease in Q,/ Q " which was reversed on turning the animals supine (Fig. 1 . 10). The improvement in gas exchange was not related to changes in FRC, regional diaphragmalic molion, cardiac output. or pulmonary vascular pressures. The au­ thors were unable to explain the mecha­ nism. The prone position is also benefi­ cial in improving oxygenalion in patients with acute respiratory failure (Douglas et aI., 1977). Jones et al. (1986), however, using two gases of markedly different dif­ fusivilies in nine normal subjects, found that the effect of posture on gas mixing represents a conveclive and diffusive-de­ pendent change in the distribution of ventilation. The effect of posture was not solely due to lung volume changes. How­ ever, this report and the two previously published human studies suggest that pa-

'" :J: E E ..

o to Q.

Time (min) Figure 1.10. Effect on PaO, of changing from the prone to supine position in dogs with oleic acid­ induced acute lung injury. Animals were kept prone for 1 5-45 min after oleic acid injection before the first measurement. (From Albert RK, Leasa D, Sanderson M, Robertson HT, Hlastala MP: The prone position improves arterial oxygenation and reduces shunt in oleic-acid-induced acute lung injury. Am Rev Respir Dis 135.£28-633. 1987.)

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

32

tient positioning should be used as a means of reducing the requirements for oxygen and PEEP during mechanical ventilation. The prone position is the postural drainage postion for drainage of the superior segments of the lower lobes. Quadriplegic patients nursed on Stryker frames may spend considerable time prone. The lack of information provided about positioning of palients during chesl physiotherapy, in many publications, makes interpretation of pulmonary func­ tion changes due to the therapy alone difficull. Use of Bronchoscopy

Perruchoud et a!. (1980) found thai bronchoscopic lavage and aspiration with saline and acetylcysteine, followed by positive pressure inflation of the atelec­ tatic lung through the bronchoscope, was successful in reexpanding atelectasis in 37 of 51 patients. Rigid bronchoscopy was used in the 1 1 spontaneously breathing patients and fiberoptic bronchoscopy in Ihe 40 mechanically ventilated patients. Eleven patients had partial reexpansion of the atelectasis and Ihree showed no change. Atelectasis recurred in three pa­ tients on the first day and five on Ihe sec­ ond, and seven patients developed pneu­ monia. Lobar atelectasis was more successfully cleared (85%) than segmen­ tal atelectasis (47%) using these tech­ niques. Segmental atelectasis also oc­ curred more frequently. All Ihe patients described had recent radiological evi­ dence of mucoid impaction. Some of the patients were apparently given physio­ therapy, but no information is provided about the success rate of reexpansion with physiotherapy alone. Bronchoscopy may be indicated if chest physiolherapy fails to reexpand an atelectasis. Bronchoscopy may be more efficacious in the non intubated patient with atelec­ tasis than in the intubated patient (Fried­ man, 1 982). There are a few indications for fiberoptic bronchoscopy after an ini­ tial aggressive regimen of respiralory therapy. Bronschoscopy may be neces­ sary when a symptomatic patient is un­ able to tolerate vigorous respiratory ther­ apy, an important diagnostic question coexists, massive collapse is u nrespon-

sive to respiratory therapy, or when pa­ tients do not show improvemenl after 24 hr of respiratory therapy (Marini et a!., 1 982). Selective positive pressure venti­ lation through the suction port of a cuffed fiberoptic bronchoscope was found effec­ tive in the expansion of refractory atel­ ectasis (Harada et a!., 1 983). Fourteen of 1 5 palienls wre successfully managed; however, in six patients atelectasis re­ curred. Bronchoscopy produces marked he­ modynamic changes when performed under topical aneslhesia. Lundgren et a!. (1982) found mean arterial pressure in­ creased by 30%, heart rate by 43%. car­ diac index by 28%. and pulmonary wedge pressure by 86% compared to pre­ bronchoscopic control values in 10 pa­ tients. The patienls were premedicated with morphine and scopolamine and the pharynx, larynx, Irachea, and bronchi were aneslhetized with 250 mg of lido­ caine. Flexible fiberoptic bronchoscopy was performed without supplemental ox­ ygen and PaO, decreased significantly from 75 ± 3 mm Hg before bronchoscopy 10 67 ± 3 mm Hg during bronchoscopy. Jaworski et a!. (1988) found that fiber­ optic bronchoscopy was no better than routine physical therapy in prevention of ateleclasis after lobectomy for lung tumor. Twenty postoperative patients were studied. Five of six patienls who produced more than 30 ml spulum per day developed atelectasis. One patient who received only physiotherapy re­ q uired Iherapeutic bronchoscopy. In the group who received routine bronchos­ copy in the postanesthesia recovery room, one patienl also required repeated bronchoscopy. There was no difference in lCU or hospital stay or duration of chest tube placement posloperatively. These findings make fiberoptic bronchos­ copy undesirable in most critically ill pa­ tients. The majority of critically ill lCU patients can tolerate the position changes necessary for chest physiotherapy (see Chapter 3). In our experience (Mackenzie et a!., 1980; Mackenzie and Shin, 1 986) and olhers (Friedman, 1 982; Marini et a!., 1 979, 1 982) bronchoscopy is only used when atelectasis persists longer than 48 hr or a realistic diagnostic dilemma exists.

CHEST PHYSIOTHERAPY. ICU CHEST PHYSIOTHERAPY. AND R ESPIRATORY CARE Summary of Literature Update Since

1 980

The importance of diaph ragmatic mal­ function as an etiology in development of respiratory complications has been estab­ lished since 1 980. The interaction of the chest wall and abdomen in the relation­ ship that leads to reduced lung volume and atelectasis has nol been as well in­ vestigated. Pain is now considered less of a causative factor in the postoperative fall in FRC than was thought before 1 980. Certainly epidural and intrathecal opi­ ates have revolutionized pain relief after surgery so that in many studies this is no longer a confounding variable (Ricks ten et al.). However. epidural and intrathecal opiates have not reduced the incidence of postoperative respiratory complica­ tions or resulted in improved diaphragm function. Epidural bupivacaine does im­ prove diaphragmatic function after upper abdominal surgery (Mankikian et al.. 1 988) but as yet has not been shown to reduce respiratory complications. A major confusion that has developed since 1 980 is in the effects of posture on gas exchange in patients with unilateral lung disease. Prone positioning. depen­ dent positioning of the good lung, and the effects of chest wall compliance are fac­ tors that may improve gas exchange ir­ respective of the therapeutic modality under study. The effects of increases in transpulmonary pressures on lung and cardiac function are numerous. The data are confusing in many of the reported studies that employ small numbers of pa­ tients. It seems physiologically unlikely that PEP is greatly different from CPAP or PEEP if they are all applied intermit­ tently. None of these means of increasing transpulmonary pressure may be differ­ ent from deep breathing. As shown so dramatically by Driks et al.. PEEP may reduce atelectasis but increase VA/Q mismatch. The mechanics of what occurs during surgery requires more investiga­ tion to find the causes of loss of lung vol­ ume and impaired diaphragmatic func­ tion after surgery. All the relatively noninvasive techniques for removal of secretions are genera l ly preferred over bronchoscopy, which produces major cardiorespiratory disturbance in the crit-

33

ically ill patient and requires physician participation. Bronchoscopy is more costly and no more effective than chest physiotherapy maneuvers for secretion removal. WHAT IS CHEST PHYSIO THERAPY?

Encompassed i n the use of the term chest physiotherapy are five maneuvers: (1 )postural drainage. (2) chest wall per­ cussion and vibration. (3) coughing. (4) suctioning, and (5) breathing exercises in the spontaneously breathing patient. Breathing exercises include the forced expiration technique or huffing. dia­ phragmatic costal excursion. and lateral costal excursion exercises. In addition, patient mobilization is used whenever possible. These maneuvers are discussed in detail in subsequent chapters. Postural drainage, manual percussion. and chest vibration are applied until specific end­ points indicate therapy should cease. The end-points include increased air entry, reduced adventitial breath sounds. increased lung/thorax compliance, ces­ sation of sputum production. or patient intolerance. Duration of therapy may. t herefore. vary from 15 to 90 min and re­ flects the extent of pulmonary dysfunc­ tion. Chest physiotherapy, including in­ structions on turning. frequency of application, methods of performance. things to avoid. and evaluation of effec­ tiveness. is described in abbreviated form in Appendix II\, WHAT ARE THE OBJECTIVES O F CHEST PHYSIO THERAPY

The objectives of chest physiotherapy include clearance of secretions from large and small airways and reexpansion of nonventilated lung. The goal of chest physiotherapy is to obtain this favorable outcome equally or more effectively than bronchoscopy without the invasiveness, trauma, risk of hypoxemia. complica­ tions. physician involvement, and cost that bronchoscopy requires (Mackenzie and Shin, 1986). A further objective of chest physiotherapy is to specifically im­ prove ventilation to areas of local lung obstruction. In this respect it differs from blow bOllles, incentive spirometry. bron-

34

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

chodilators, mucolytic agents, IPPB, pres­ sure support, or application of positive pressure (CPAP) that are applied to both lungs indiscriminately. Chest physio­ therapy aims to aIter local lung expan­ sion and is physically directed at the local lesion. Postural drainage places the local lesion in the ideal position for grav­ ity to promote drainage of secretions. Percussion and vibration over the skin surface area of the local lesion assist pos­ tural drainage and are also directed at the lung wilh the local lesion. In addition. one of the objects of chest physiotherapy is to produce benefit when infiltrates are generalized. If the objectives of the chest physio­ therapy are achieved, an increase in local lung expansion should occur and a par­ allel increase in perfusion to the affected area would result . If secretions are cleared from larger airways, airway re­ sistance and now obstruction should de­ crease. Clearance of secretions and im­ proved ventilation of small airways should increase lung compliance. If clearance of secretions from both large and small airways occurs, it is reasonable to assume that the work of breathing and oxygen consumption should decrease, and gas exchange improve. Furthermore, if these objectives are achieved, the in­ cidence of postoperative respiratory in­ fection, morbidity, and hospital stay for those with acute and chronic lung dis­ eases should be reduced. The mecha­ nisms by which chest physiotherapy at­ tempts to achieve these objectives are discussed in Chapter 7, pp. 237-242.

fore. supplements chest physiotherapy treatments on evening and night shifls. Nurses can also help by providing a backup when the case load becomes too great for the number of physical therapists. The essence of application of chest physiotherapy to these patients with acute lung problems lies in a cooperative approach that includes the critical care physician, radiologist, physical therapist, and nurse. Patienls receive a daily por­ table anteroposterior chest x-ray that is reviewed in the morning critical care conference by the radiologist, following the palient case presenlation. At least one physical therapist attends the daily morning rounds. An attempt is made by the radiologisl to identify not jusl lobar involvement but, more particularly, seg­ ments within a lobe. This approach, even with potential sources of error, has a rel­ atively high success rate (Ayella, 1978). The chest physical therapist. therefore, approaches the patient wilh knowledge of the early morning chest x-ray and a verbal reporl that enables the patient to be positioned with the affected segmenl uppermost. After chest physiotherapy, a follow-up chest x-ray is requested on a patient with a complete lobar atelectasis. Treatment is given unlil clinical signs of improvement are noted and as long as sputum is obtained. Details of the exact procedures and Ihe outcome are dis­ cussed in laler chapters and are summa­ rized in Appendix Ill.

CHEST PHYSIOTHERAPY ORGANIZATION

Physiolherapy should not be consid­ ered in isolalion from overall respiratory care managemenl. All the data in me­ chanically ventilated palients presented in this book before October 1978 refer to ventilation Ihat was completely con­ trolled by means of a time-cycled vol­ ume-preset mechanical ventilator, the Engslrllm 300. Since 1978, intermittent mandatory ventilation, high-frequency ventilation. CPAP, pressure supporl ven­ tilation, independent lung ventilation. continuous now ventilation, and combi­ nations of high-frequency and continu­ ous now ventilation and conventional positive pressure ventilation were used.

The number of palients treated with chest physiotherapy at our institution has risen annually. The statislics for Ihe types and numbers of patients treated with chest physiolherapy appear in Ap­ pendix I. The patient population is sum­ marized in Appendix I. There are 16 full­ time physical therapists who provide not only chest, but also rehabilitation care. The physicial Iherapists train the nurses by means of in-service tutorials. Audio­ visual aids assist in this process (Macken­ zie el al .. 1978). The nursing staff, Ihere-

RESPIRATORY MANAGEMENT

CHEST PHYSIOTHERAPY. ICU CHEST PHYSIOTHERAPY. AND RESPIRATORY CARE

35

To prevent unnecessary complications. minimize confusion resulting from dif­ ferent techniques and ideas used by other training centers, and aid in collec­ tion of meaningful data, respiratory care may be standardized. The following suggestions for respiratory care are. therefore, dogmatic and certainly are not necessarily applicable in all sit uations.

chanica I ventilation are no longer required in a patient with a tracheos­ tomy. the patient is extubated. Use of fenestrated tracheostomy tubes and pro­ gressive reduction in tracheostomy tube size needlessly prolong tracheal intuba­ tion and efforts to decannulate. Both the fenestrated and small tracheostomy tube increase airway resistance and the work of breathing and decrease cough effec­ tiveness and secretion clearance (Criner et aI., 1 987).

Intubation

Mechanical Ventilation

Indications for Intubation, Ventilation, Weaning, and PEEP

The indications for intubation include 1. Airway obstruction that cannot be simply relieved 2. PaO, of less than 80 mm Hg on supple­ mental 0, or of less than 60 mm Hg on room air 3. Patient in shock with a systolic blood pressure of less than 80 mm Hg 4. Severe head injury or unconscious­ ness 5. Anticipated surgery. In an emergency patients are preoxy­ genated, and cricoid pressure (Sellick, 1961 ) is applied to prevent regurgitation of gastric contents during rapid sequence intubation. with thiopental (when indi­ cated) and succinylcholine. Intubation is initially carried out by the orotracheal route. If the patients are incapable of protect­ ing their own airway due to unconscious­ ness. incompetent laryngeal reflexes. or upper airway obstruction. then. after about 1 0-14 days a tracheostomy is usu­ ally performed, and a cuffed tracheos­ tomy tube is placed. Patients with severe head injury and spasticity may have a tracheostomy much earlier. within 1 -3 days of ICU admission. Since tracheal tubes promote secretion production and reduce the effectiveness of coughing, ex­ tubation is carried out once the tube is no longer indicated to protect the airway. and the patient meets weaning criteria. Therefore, patients would not normally breathe spontaneously for prolonged per­ iods through orotracheal or nasotracheal tubes; instead. a tracheostomy is per­ formed or the patient is extubated. Simi­ larly when airway protection and me-

Mechanical ventilation is used during tracheal intubation unless the patient is about to be extubated or spontaneous respiration with PEEP is employed. For completely controlled mechanical venti­ lation minute ventilation of the ventila­ tor is adjusted to satisfy respiratory drive. so that the patient is kept in phase. This commonly means hyperventilation to a PaCO, of 30-35 mm Hg. Ventilation is most frequently kept at a rate of 1 2-20 breaths/min. and a tidal volume in the range of 10-15 mljkg is used. Many cen­ ters use a high tidal volume and low res­ piratory rate (Bendixen et al.. 1 963). The patient with a head injury is routinely hyperventilated to a PaCO, of 25-30 mm Hg as one part of the therapeutic maneu­ vers to reduce intracranial pressure. Other maneuvers for this purpose may include steroids, di uretics, barbiturates. and monitoring of intracranial pressure. If the patient is out of phase with the ventilator. pneumothorax. inadequate minute ventilation, retained secretions and inadequate sedation should be con­ sidered. investigated. and if present. cor­ rected. Narcotics and benzodiazepines or both are used for sedation and pain relief. Muscle relaxants. such as vecuronium or pancuronium. are used only if the cause of asynchrony cannot be found and cor­ rected. The two groups of patients in whom muscle relaxants are used most frequently are decorticate or decerebrate patients with head injury and those who develop generalized sepsis. Patients with severe head injury and elevated intracra­ nial pressure may also receive high doses of barbiturates.

36

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Weaning from Mechanical Ventilation

The weaning criteria that are used include 1. Partial pressure of arterial oxygen/ fractional inspired oxygen (PaO,/ FlO,) greater than 250 2. Vd/V, less than 0.55 (by nomogram or direct measurement) 3. Maximum inspiratory force greater than -20 cm H,O 4. Vital capacity 1 5 ml/kg body weight or greater than 1 000 ml 5. Total lung/thorax compliance greater than 30 ml/cm H,O 6. No muscular fatigue, neurological or nutritional indication for continued mechanical ventilation. If all the weaning criteria are met, the patient may be placed on a T-piece or tra­ cheostomy collar. A T-piece is normally used when the patient has an orotracheal or nasotracheal lube. Arterial blood gases are measured after 30 min and 1 hr of spontaneous breathing. If the patients maintain adequate arterial blood gases, extubation is indicated. Gradual weaning is not used. As an alternative to the T­ piece, gradually decreasing rates of IMV, CPAP, and pressure support may be em­ ployed. This approach considerably in­ creases the duration of weaning but has many advocates. The indications for reinstitution of me­ chanical ventilation include Fall in PaO, below 60 mm Hg when breathing 10 liters/min of 40% 0, by an aerosol humidifier 2. Increase in PaCO, of 10 mm Hg/hr, or 5 mm Hg/hr for 3 consecutive hours 3. Respiratory rate consistently exceed­ ing 50/min 4. Sudden deterioration of conscious­ ness. 1.

PEEP

PEEP may be applied if PaO,/FIO, is less than 200. However, there are many critical care physicians who advocate use of PEEP of at least 5 cm H,O in all me­ chanically ventilated patients. Initially, 5 cm H,O is applied. and arterial blood gas analysis is repeated after about 1 5 min. If

PaO,/FIO, is still not in excess of 200. chest physiotherapy may be indicated if there is retention of lung secretions. If, following chest physiotherapy, the ratio is still not in excess of 200, further incre­ ments of PEEP are added to a maximum of 10 cm H,O. If, with 10 cm H,O. PEEP, and chest physiotherapy, the PaO,/FIO, is still not greater than 200, a Swan-Ganz flow-directed, thermistor-tipped, pulmo­ nary artery catheter is inserted. Further increases of PEEP above 10 cm H,O may be detrimental to cardiorespiratory func­ tion and should be monitored with arte­ rial, mixed venous gas, and cardiac out­ put determinations. Normally PaO, is maintained at about 100 mm Hg by adjustment of FlO, and PEEP. FlO, is not normally increased above 0.6. Exceptions may occur with carbon monoxide poisoning. chest phys­ iotherapy in an extremely unstable pa­ tient, hyperbaric oxygen therapy, and dif­ fusion problems in the lung. Alternative Modes 01 Mechanical Ventilation

Intermittent Mandatory Mechanical Ventilation

Intermittent mandatory ventilation (IMV) was introduced in 1 973 by Downs et a!. Compared to conventional inter­ mittent positive pressure ventilation, the essential difference is provision of a par­ allel inspiratory gas circuit that allows the patient to breathe spontaneously be­ tween mechanical breaths. The advan­ tages and disadvantages of intermittent mandatory ventilation are summarized by Benzer (1982) and Willatts (1985a,b). The advantages claimed for IMV include reduced sedation requirements, sponta­ neous breathing that may prevent respi­ ratory muscle atrophy, improved auto­ regulation of acid-base balance, reduced mean intrathoracic pressure, reduced risk of barotrauma, and an improvement in renal excretory function. The claimed benefits of earlier weaning are not sub­ stantiated by clinical trials (Weisman et a!., 1983). Modification of simple IMV cir­ cuits is necessary to reduce the work of breathing and respiratory muscle fatigue.

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

.Airway pressures during IMV may be equally high as with conventional me· chanical ventilation. High-Frequency Ventilation. High·fre­ quency ventilation (HFV) utilizing low tracheal pressures can maintain gas ex­ change while reducing the effects of pressure on the lung and circulation. HFV is indicated in bronchopleural fis­ tula and necrotizing pneumonia. HFV may be beneficial in respiratory failure unresponsive to conventional tech­ niques, although there is little agreement in the literature on benefits of HFV com­ pared to conventional mechanical venti­ lation. High frequency jet ventilation at 60-300 breaths/min (bpm) produces gas exchange by modifications in gas mixing and streaming. Air trapping occurs at higher frequencies. High-frequency os­ cillation has the advantage of an active expiratory phase that prevents gas trap­ ping and allows frequencies of 1000-3000 bpm to be achieved using small tidal vol­ umes. It may be clinically valuable to combine HFV with conventional me­ chanical ventilation (Nunn, 1 987). Differential Lung Ventilation (DLV). In circumstances where disease or surgery result in pathological changes in one lung not found in the other, DLV may be ben­ eficial. Use of two synchronized ventila­ tors enables ventilation and PEEP to be adjusted independently. A double-lumen cuffed tube is used to separate ventila­ tion to the right and left lung. DLV has limited long-term use because double­ lumen tubes are large and traumatize the larynx, trachea, and particularly the mainstem bronchi. Continuous Flow Ventilation (CFV)

Apneic oxygenation (AO) and tracheal insufflation of oxygen (TRIO) were de­ scribed by Hirsch (1905) and Meltzer and Auer (1909). More recently Lenhert et al. (1982) described a technique of endo­ bronchial insufflation (EI) of flows of 1 liter/kg/min. AO, TRIO, and EI are all CFV techniques and they produce gas ex­ change without any chest movements. EI differs from the others in producing nor­ mal oxygenation and CO, excretion for up to 5 hr with room air insufflation. Both

37

AO and TRIO require a, and do not ex­ crete adequate amounts of CO, for pro­ longed use without development of res­ piratory acidosis. CFV techniques may be used as adjuncts to conventional and DLV techniques. The main clinical use of CFV occurs during extracorporeal mem­ brane CO, removal. CFV may be a useful adjunct to augment oxygenation (Gatti­ noni et aI., 1 980). Inverse Ratio Ventilation

This technique aims to increase the relative duration of inspiration but may cause gas trapping and elevated intratho­ racic pressure. Depending on the respi­ ratory pathology. reversal of the normal I : E ratio of 1 : 2 to 3 : 1 may decrease intra­ pulmonary shunt, dead space, and PaCO, (Perez-Chada et aI., 1 983). If decreased lung compliance occurs with normal air­ way resistance, time constants for inspi­ ration are reduced. In those patients, pro­ longation of inspiration by inverse ratio ventilation may improve gas exchange by increasing the lime that otherwise closed alveoli are held open (Willatts, 1985b). Exclusions in Respiratory Care

No assist modes or sigh mechanisms were used during mechanical ventilation of any of the patients treated at our insti­ tution. No IPPB machines were used for delivery of bronchodilators or mucolytic agents to assist secretion clearance and no inhaled drugs, other than water vapor, were used in any of the patients treated before 1 980. All patients received humid­ ification from the moment of intubation, through anesthesia, and during any sub­ sequent mechanical ventilation. No spontaneously breathing patients were treated with blow bottles. Tracheal la­ vage was rarely part of routine patient care. Secretions were removed in intubated patients by means of postural drainage, percussion, vibration or shaking therapy, and tracheal suctioning. Mobilization was encouraged in all patients, and coughing, in those who were spontane­ ously breathing. Lung hyperinflation or "bagging" was not employed when chest

38

CHEST PHYSIOTHERAPY I N THE INTENSIVE CARE UNIT

physiotherapy was given to mechanically ventilated patients. Breathing exercises were used in spontaneously breathing patients. Rarely was bronchoscopy used. Chest physiotherapy replaced bronchos­ copy as the first treatment of choice for removal of retained secretions at our in­ stitution in 1 974. The number of thera­ peutic bronchoscopies between 1 972 and 1979 is shown in Table 1 .2. Bronchoscopy is not used when an air bronchogram is visible on chest x-ray to peripheral lung. Bronchoscopy is employed for therapeu­ tic effect when 24 hr of chest physiother­ apy has failed to reexpand atelectatic lung and a diagnostic dilemma exists. De­ tails of procedures followed in applica­ tion of chest physiotherapy are discussed in subsequent chapters and summarized in Appendix III. MISCONCEPTI ONS ABOUT EFFECTS OF CHEST PHYSIOTHERAPY

Misconceptions exist about the use of chest physiotherapy in patients with acute lung pathology. Many reports in the literature discuss the use of chest physiotherapy in patients with chronic lung diseases, such as bronchiectasis, cystic fibrosis. or bronchitis. These pa­ tients may not be hospitalized and often receive treatment at a clinic or on an out­ patient basis. They have chronically ab­ normal lungs. Therefore, it may not be justified to extrapolate what occurs fol-

lowing chest physiotherapy in these pa­ tients to those who have acute lung dis­ ease in otherwise normal l ungs. The patient population studied also in­ fluences the likely outcome of maneu­ vers designed to remove retained lung secretions. For example, the response fol­ lowing chest physiotherapy of a patient with atelectasis secondary to pain from a stab wound in the chest is more likely to be favorable than is the response to chest physiotherapy of a patient with adult res­ piratory distress syndrome and multisys­ tem blunt trauma. When comparisons are made between two forms of therapy de­ signed to remove secretions from the tra­ cheobronchial tree, a single use of the therapy should be compared on similar patient populations with similar lung pa­ thology. Confounding variables such as differences in position during therapy may produce changes in oxygenation that are unrelated to the therapy. Apples must not be compared to oranges. When chest physiotherapy is compared to any other therapy, both therapies must be performed on a similar patient popula­ tion. This is not always done. Therefore, confusion may exist concerning effec­ tiveness of the therapy in improving the patient. This, of course, brings up a most important point. What is meant by improvement? There is difficulty in adequately quan­ titating the effectiveness of chest phys­ iotherapy. Subjective claims that the pa­ tient feels better. that clinical signs

Table 1.2 The Number of Patients Receiving Therapeutic Bronchoscopy for Retention of Lung Secretions for 1 972-1980 Is Compared to the Total Number of Admissions' Year

Total Admissions

Number of Bronchoscopies

Number of Patients Requiring Bronchoscopy

1 972

61 5

31

19

1 973 1 974 1 975 1 976 1977 1 978 1 979

982 872 920 1 , 1 05 1 ,023 1 ,053 1 ,249

21 7 9 14 9 8 18

17 7 9 12

'The data for 1 972-1975 to MIEMSS.

were kindly supplied by J.

I

6 14

Comments One patient had 7 Six patients had 2 Four patie nts had 2

Two patients had 2 Two patients had 2

Two patients had 2

One patient had 3 Two patients had 2

Hankins, M.D., Thoracic Surgical Consultant

CHEST PHYSIOTHERAPY, ICU CHEST PHYSIOTHERAPY, AND RESPIRATORY CARE

improve, or that the chest x-ray appear­ ance is more favorable are very difficuli to interpret. Objective evidence is sparse, and it was agreed, in the excellent con­ ferences sponsored by the American Thoracic Society in 1974 (National Heart and Lung Institute) (NHLI) and in the similar conference (National Heart, Lung and Blood Institute) (NHLBI) in 1 979 (see pp, 245-248 for summary), that there are very few, adequate, and acceptable ways reported that provide evidence of the ef­ fectiveness of chest physiotherapy in chronic lung diseases, Lung function tests would appear to be the best param­ eter for comparison, However, even they may be prejudiced either by requiring in­ vasive and sophisticated techniques for measurements, such as xenon inhalation, or by the tests themselves allering the pa­ rameter that they purport to measure, For example, the maneuver of forced ex­ piration is known to cause small airway closure and may induce bronchospasm in some patients (Nunn et aI., 1 965), FEV, measurement may, therefore, not accu­ rately reflect the effect of chest physio­ therapy, as some authors have suggested (Campbell et al.), Another important rea­ son why confusion may exist about the use of chest physiotherapy is the wide range of therapies that are described, Claims that chest vibration did not pro­ duce a statistically significant difference in clearance of sputum when compared to coughing alone must be questioned when it is learned that the therapy was performed in the silting position with a mechanical vibrating pad (Pavia et al.), In our experience, upper lobe secretion re­ tention is uncommon (see Appendix I), Far more frequently, lower lobe collapse is seen, especially left lower lobe col­ lapse, Therefore, the silling position is inappropriate for drainage of lower lobe secretions, If chest vibration was per­ formed by a physiotherapist, with the pa­ tient in the appropriate postural drai nage position, until there was clinical evi­ dence of secretion clearance, the out­ come may be different. In other reports, "chest physiotherapy" includes suctioning but no percussion, manual vibration or postural drainage (Fox et al" 1 978), Some physiotherapy techniques include bagging (where the

39

lungs are hyperinflated), and vibration or chest shaking is applied during expira­ tion (Gormezano and Branthwaite, 1 972a), The duration of therapy also shows an amazingly wide variation, This reasonably may be expected to produce different resulis, In some centers, ther­ apy lasts only a few minutes (Hedstrand et al.), In other centers, it may be applied for 7-20 min (Gormezano and Branth­ waite, 1 972a), or in others the therapy is restricted to a predetermined duration of 1 0- 1 5 min (Newton and Bevans, 1 978), Some reports also include the adminis­ tration of nebulized solutions by IPPB as part of chest physiotherapy (Graham and Bradley, 1978), Therefore, a reported 20 min of chest physiotherapy may actually consist of 1 5 min of IPPB and only a short period of postural drainage percussion and coughing, All these variables must be appreciated before drawing conclu­ sions from the literature, I I seems reasonable to assume that un­ less sputum is produced by chest phys­ iotherapy, the treatment is not likely to be beneficial. II is, however, doubtful that the more sputum produced, the greater the benefit. Sputum removal may cause a fall in PaO, and an increase in in­ trapulmonary shunt because of alleration in ventilation perfusion relationships, This was suggested as the probable cause for the falls in PaO, noted by Gormezano and Branthwaite ( 1972b) following chest physiotherapy in chronic lung disease patients, Partial reexpansion of a lobar atelectasis may reverse the vascular compensating mechanisms that previ­ ously reduced intrapu lmonary shunt. Most clinicians agree with Murray that sputum production is essential if chest physiotherapy is to be effective for the treatment of chronic lung disease, How­ ever, the removal of 30 ml. that he states as the minimum necessary to produce benefit, is excessive in the patient with normal lungs and acute secretion reten­ tion, If sputum is removed from the smallest airways. less should produce beneficial effects (see p, 220), Sputum volume measurement is noto­ riously unreliable (Bateman et al.), The expectorated measurement has consid­ erable sources of error, as the patient can swallow sputum. so reducing the volume

40

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

collected, and equally, the patient can add saliva apparently increasing the quantity. The effectiveness of humidifi­ cation and patient hydration may also af­ fect sputum volume. Except for our data presented in Chapter 7, few studies mea­ sure volume of sputum production in tra­ cheally intubated patients. The cuffed tracheal tube removes two of the major sources of error in sputum collection mentioned above. Volume of sputum may not be the primary consideration for assessing chest physiotherapy effective­ ness for the ICU patient. In our experi­ ence treating acute lung pathology, there is lillie relationship between the volume of sputum produced in excess of 5 ml and the benefits in terms of improved physi­ ological parameters. In acute processes, as in chronic lung disease, different lung problems produce different volumes and types of sputum. The lung contusion pro­ duces mostly blood, not sputum, yet chest physiotherapy may improve venti­ lation/perfusion relationships (Macken­ zie and Shin, 1 979a). Lobar and segmen­ tal atelectasis may clear when only a small volume of sputum is removed, yet dramatic improvement in intrapulmo­ nary shunt and total lung/thorax compli­ ance may occur on reexpansion. In pa­ tients with established pneumonia, lillie improvement may be apparent whether a large or a negligible amount of sputum is produced. The long-term benefits of re­ moval of retained sputum are unknown. It is thought that it does not alter the course of chronic lung diseases (NHLI, 1974), but in acute lung diseases, espe­ cially in an ICU environment, it seems likely that it would reduce the incidence of respiratory tract infection and, there­ fore, morbidity (and possibily mortality). Areas of dispute concerning the compo­ nents, alternatives and disease processes treated by chest physiotherapy are sum­ marized in Table 1 .3. This table also pro­ vides our opinion on some controversial points. SUMMARY

Objective studies of chest physio­ therapy are few and the techniques em­ ployed are not standardized or ade­ quately described. The patient popula-

tions studied are variable, sometimes inappropriately excluding those patients likely to or including those least likely to benefit. Chest physiotherapy is used frequently in pediatric ICUs, surgical ICUs for res­ piratory problems occurring after sur­ gery, and for treatment of chronic lung disease. In the acute selling. physiother­ apy appears equally as good as bronchos­ copy for reexpansion of atelectasis. For patients with chronic lung disease it seems effective when production of se­ cretions is great. Some common misconceptions about chest physiotherapy that occur because of the lack of standardization were em­ phasized. Chest physiotherapy tech­ niques were briefly described. Controlled mechanical ventilation was considered the treatment of choice for the acute management of the severely in­ jured and pathological lung. Removal from mechanical ventilatory support does not need to be gradual in patients recovering from acute lung pathology. Changes in lung/thorax compliance were thought to be a good indicator of the benefits and end point of chest phys­ iotherapy. Assist modes of ventilation, sighs and hyperinflation, bronchodilator and mu­ colytic agent inhalations, and use of blow bottles and incentive spirometers have clouded the interpretation of the results of chest physiotherapy. Therapeutic bronchoscopy is rarely required for spu­ tum removal when chest physiotherapy is employed. It is clear from the literature since 1980 that posoperative pain is not the major factor in development of res­ piratory complications. Diaphragmatic dysfunction after upper abdominal sur­ gery appears to be an important etiology for respiratory complications that is re­ versed by epidural bupivacaine but not by spinal or epidural opiates. Despite a multitude of mechanical aids to lung ex­ pansion such as PEEP. CPAP, PEP, pres­ sure support, and inverse ratio ventila­ tion none is convincingly superior to chest physiotherapy, deep breathing, and position changes in prevention of respi­ ratory complications in the ICU. For me­ chanically ventilated patients, the re­ cently described techniques to improve

() ::t m

Table 1 .3 Conflicting Data and Points 01 Contention Concerning Chest Physiotherapy For

�

Against

Authors Opinion and Practice

Component May not add to the effect of cough and suction"'�

Percussion and vibration

Assists secretion clearancel-3

Postural drainage (PD)

Peripheral clearance of secretions is enhanced with PD.' Significantly more sputum is produced with PD and the forced expiratory technique.' PD without positive expiratory pressure (PEP) produced more sputum than PEP therapy in the sitting position.' The addition of PD and coughing improves mucociliary clearance9

Both cough and exercise are superior to PO at clearance of secretions.'O Sputum production is reduced with CPT including PD compared to FET or PEP without PD." PD is no better than cough alone' 2.13

CPT does not add benefits to cough alone,I2 80th clear central and peripheral airways'

CPT but not cough produces peripheral clearance of secretions.' Cough effects are limited to the central airways, proximal to the fifth generation.ls Repetitive coughing may cause fatiguel6 and bronchospasml1

Cough

Forced expiratory technique (FET) and PEP

FET is more effective than cough alone. FET with PD produced more sputum than CPT.' PEP prevents atelectasis after upper abdominal surgery." FET is better tolerated than CPT"

Sputum clearance was less effective when PEP is included in treatment. PEP clears less secretions than PD alonee

Assists secretion clearance; mode of action postulated in Chapter 4 and on p. 240-242 Identification of the involved lung segment is essential to determine which of 1 1 different positions is correct for gravity assisted segmental drainage,'" PO has an additive effect to percussion, vibration, coughing, and FET. PD improves central airway clearance when cough is impaired It is inappropriate to separate cough from other CPT techniques. Cough occurs spontaneously and concurrently when secretions are loosened or suctioned. Cough is important for removal of secretions advanced to the central airways. Protracted coughing should be avoided FET is not shown to improve mucus clearance in peripheral airways. May be helpful to clear central airways. Physiological basis unknown.19 PEP produces only short-lived benefits

." ::t -< (J)

�

::t

m :D »

." .-< o c () ::t

m

(J) ..... ." ::t -< (J)

� ::t

m :D » ."

:
(5 .... I m JJ » .., -< z .... I m z .... m z U> ?ww) K L Figure 3.3. (Continued) (G) The posterior segment of the right upper lobe is drained with the patient positioned one-quarter turn from prone. The bed remains flat. (H) Postural drainage of the right upper lobe posterior segment in a patient with a head injury, left pulmonary contusion. frac· tured ribs, left pneumothorax, and pelvic and limb fractures. He received physiotherapy following splenorrhaphy and internal fixation of bilateral femur fractures. (I) Upper: right middle lobe position; lower: lingula position. The right middle lobe and lingula are drained with the patient one·quarter turn from supine. A 12-inch bed elevation is recommended. (J) A patient receiving postural drainage of the lingula. (K) The bed is flat and the patient is prone to drain the superior segments of both lower lobes. (L) A patient with mUltiple-system involvement, including cervical spinal cord, chest and abdominal injuries.

POSTURAL DRAINAGE. POSITIONING, AND BREATHING EXERCISES

99

M

Q� Figure 3.3. (Continued) (M) The lateral segment of the right lower lobe is drained with the patient lying on the left side. The foot of the bed is elevated. (N) Postural drainage of the lateral segment of the right lower lobe is shown in a patient who sustained a head injury. a fractured left femur. fractured right lateral malleolus and fibula. ruptured spleen. and liver lacerations. Pancuronium. a neuromuscular blocking agent, was used because the patient was agitated and out of phase with the ventilator. The patient has resting hand splints to preserve functional range of motion. Neufeld traction permits turning to the left side. (0) The right side-lying head-down position is used for postural drainage of the lateral segment of the left lower lobe and the medial segment of the right lower lobe. (P) The lateral segment of the left lower lobe is posturally drained in a patient with bilateral femoral fractures, a fractured left tibia, left lung contusion. and a chest tube draining a left hemothorax. Neufeld traction and exoskeletal fixation devices allow the patient to be turned. A Philadelphia collar IS being worn Since the seventh cervical vertebra could not be visualized on lateral cervical spine x-ray. (a) The posterior segments of both lower lobes are drained in the prone head-down position. (R) This patient had a right pneumothorax, lung contusion and soft-tissue lac­ erations around the right elbow. He also had a laparotomy for repair of liver lacerations and mes­ enteric tears. (S) The supine head-down poSitIOn is used to drain the ante nor segments of both lower lobes. (T) This patient with multiple extremity fractures receives postural drainage of the anterior segments of the lower lobes.

100

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Figure 3.4. Two people are required to pull a patient with multiple injuries to the side of the bed.

3. Move the patient to the side of the bed before turning (Fig. 3.4). If the patient cannot assist, two people slide the pa­ tient to the side of the bed. One person lifts the thorax while the second moves the hips. 4. Move lines and electrocardiogram wires away from the side onto which the patient is turning. 5. While facing the patient, place one hand over the shoulder and the other over the hip to rotate the patient onto the side. A second person may be re­ quired to move the hip and shoulder back so that the patient remains cor­ rectly positioned (Fig. 3.5) . A roll may be used to prevent the patient from rolling supine. Rolls, which can be made from folded sheets or blankets, are better than pillows because they are less easily compressed (Fig. 3.6).

Figure 3.5. On e therapist places a hand over the shoulder and hip; a second pulls the hips back to position the patient on his side.

Figure 3.6. A roll made from sheets (arrow) is very practical and can be used to maintain the patient in a side-lying position. Turning the Patient into the Prone Position

The following guidelines have been found useful for turning the patient into the prone position: 1 . Flex the dependent shoulder 1 80' or position the dependent arm so that the patient can be rolled over it. One ther­ apist raises the trunk while the second moves the arm and shoulder from under the patient. Patients with oro­ tracheal or nasa tracheal tubes can maintain the prone position as long as cervical rotation is not restricted (Fig. 3 . 7 ). Sometimes it may be necessary to prop the patient's head on a pillow or rolled towel to prevent kinking of the tracheal and/or ventilator tubes.

Figure 3.7. Three people are required to turn a patient with a tracheal tube prone. The de­ pendent arm is moved under the patient's tho­ rax and positioned comfortably. A roll is then placed under the chest to prevent the tracheal tube from becoming occluded.

POSTURAL DRAINAGE, POSITIONING, AND BREATHING EXERCISES

101

5. Once the patient is turned into the ap­ propriate position, the ventilator tub­ ing and monitoring equipment are checked and readjusted as necessary. Turning the Patient with Intravascular Lines Central Intravenous Subclavian Lines

Figure 3.B. A massively fractured extremity requires careful positioning with turning.

2. Patients who have orthopedic injuries or pain that inhibits turning onto the dependent shoulder may be turned prone over the noninvolved upper extremity. 3. Patients with skeletal traction or ex­ ternal skeletal fixation on the lower extremity require a third person for prone positioning. The extra person is required to position the involved ex­ tremity carefully, avoiding motion that is detrimental to the injury (Fig. 3.8). 4. Mechanically ventilated patients with tracheostomies also require the assis­ tance of three people for turning into the prone position. Two are required to raise the patient's upper thorax, while one places a roll under the chest to prevent occlusion or pressure on the tracheostomy tube (Fig. 3.9).

Figure 3.9. Placing a roll under the chest al­ lows a patient with a tracheal tube to be turned three-quarters prone. There is then adequate space for the tracheal and ventilator tubes.

Suturing of central intravenous subcla­ vian lines so that the external tubing runs parallel to the patient's thorax al­ lows full shoulder movement. The 90" side-lying position, the prone position, and full shoulder horizontal adduction can then be obtained without the line being pulled out or kinked (Fig. 3.1 0). Peripheral Intravenous Lines

Peripheral intravenous lines usually do not i nterfer with turning. Preferably, in order to allow beller mobilization, they should not cross a joint.

Figure 3,10. (A) Tubing of a subclavian line may become kinked when the patient lies on the same side. (8) Suturing subclavian lines parallel to the patient's chest allows adequate side-to-side turning and proper positioning for postural drainage.

102

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Arterial Lines

Radial lines are maintained in good po­ sition with a secure dressing. In addition, an armboard or splint may help avoid displacement. Femoral lines function in all the required positions for postural drainage, including sitting. Occasionally, hip flexion affects line function. The hip is then further extended until the proper wave form is again observed on the mon­ itor. The prone and supine positions are ideal for maintaining a good femoral ar­ terial line wave form. The presence of umbilical catheters in neonates should not be a contraindication to prone positioning. Turning the Patient with Chest, Tracheal, Feeding, and Sump Tubes Chest Tubes

Chest tubes are placed for treatment of hemopneumothorax and to adeqately drain the pleural cavity in cases of pa­ tients with empyema. They are routinely connected to an underwater seal. Suction is often added for adequate expansion of pneumothorax or to increase drainage of the pleural cavity. Kinks or compresson of the tubes are avoided during turning by careful patient positioning (Fig. 3 . 1 1 ). Turning and positioning the patient may increase pleural drainage from depen­ dent lung regions. Tracheal Tubes

Tracheal tubes may obstruct with prone patient positioning for postural drainage. I n order to prevent this prob-

Figure 3.1 1. A patient with chest tubes may be positioned prone. The tubes are carefully positioned without ten s ion or kinking.

lem, a large roll is placed under the upper thorax. This allows adequate space for the tracheostomy tube and ventilator tubing (Fig. 3.7). In order to minimize tra­ cheal trauma, the patient may be discon­ nected from the ventilator while being turned. Patients with large intrapulmo­ nary shunts that require high fractional inspired oxygen concentration (FlO,) and positive end-expiratory pressure to maintain adequate arterial oxygenation are an exception to this general rule. In these patients, instead of disconnection, an additional person assists with turning b y manipulating the ventilator tubing. This regimen of turning the intubated and mechanically ventilated patient has been found most satisfactory. Patients with tracheal tubes may be positioned prone routinely when this position is clinically indicated for chest phYSIOther­ apy. To prevent agitated patients from re­ moving their tracheal tube, hand re­ straints or sedation are used before treatment and the tracheal tube is well secured. Feeding Tubes

Patients in the ICU may receive nutri­ tional support intravenously or through orogastric. nasogastric, or gastrostomy feeding tubes. It is important for the cli­ nician altering a patient's position to be aware of the type of feeding an individ­ ual patient is receiving. Bolus feedings usually require coordination of physio­ therapy treatment with the feeding schedule. Chest physiotherapy is given before or 30 min after a feeding. Contin­ uous feedings can be stopped for physio­ therapy treatment and resumed when the patient is no longer positioned head­ down. Gastroesophageal reflux, a malfunction of the distal esophagus causing regurgi­ tation of stomach contents into the esophagus, may be associated with aspi­ ration of gastric contents into the lungs. This condition is seen in infants and chil­ dren and presents a dilemma to the ther­ apist performing chest physiotherapy. Nutritional support must be coordinated with prevention and treatment of aspira­ tion. The recommended medical treat­ ment is the semi erect position (head ele­ vated 30-60°) (Herbst and Myers, 1981;

POSTURAL DRAINAGE. POSITIONING. AND BREATHING EXERCISES

Ocha. 1981 ). especially 30-45 min after feedings. Chest physiotherapy is indi­ cated to areas of suspected or confirmed aspiration. Treatment should be coordi­ nated with feedings and administered ei­ ther before meals or no sooner than 1 hr after meals. to minimize chances of re­ gurgitation [De Ceasare. 1985). Position­ ing should be done judiciously and the patient's response to position changes should be closely monitored. If increased vomiting is noted after treatment. thera­ peutic positioning to minimize reflux. or a modified postural drainage position may be required. Therapeutic position­ ing for children with recurrent reflux is described in the literature. The prone up­ right position with partial neck flexion [Hewitt. 1976: Zimmerman and Oder. 1981 ) is thought to promote drainage of regurgitated secretions from the mouth and eliminate aspiration into the lungs. The right side-lying position may be ben­ eficial to increase gastric emptying [Wood. 1979). A prone Iyer may be used successfully in i nfants with severe cen­ tral nervous system dysfunction associ­ ated with gastroesophageal reflu x and as­ piration. The infant is placed in the prone Iyer head up 30-45 min prior to feedings. Tracheal aspiration and the need for rou­ tine chest physiotherapy treatment may be eliminated [1m Ie. 1983). The goal of treatment for patients with gastroesoph­ ageal refl ux is to provide adequate nutri­ tional support and pulmonary hygiene. Each patient should be individually as­ sessed to determine if therapeutic posi­ tioning is adequate. If adventitial breath sounds and clinical signs of chest infec­ tion persist chest physiotherapy is indicated. Sump Drains

Intraabdominal sump drains do not in­ terfere with patient pOSitioning. Care should be taken not to pull or disconnect intraabdominal tubes when moving the patient. Dislodging the sump may lead to hemorrhage or peritonitis. Turning the Patient with a Urinary Catheter

Urinary catheters do not interfere with patient positioning. It is important to

103

clamp the tubing between the collection bag and patient before turning patients. This prevents drainage of stagnant urine into the bladder. which may lead to uri­ nary tract infection. The collection bag is moved to the side of the bed that the pa­ tient faces after turning. To promote urine drainage the collection bag should be kept i n the dependent position. moved to the head or foot of the bed. depending on the patient's head-up or head-down position.

PATIENTS WITH HEAD INJURY

The two major problems of performing chest physiotherapy in patients with head injury are turning and obtaining the head-down position. Turning the Patient with Head Injury Use of Intracranial Pressure (ICP) Monitoring

ICP measurement by a subarachnoid screw or intraventricular catheter is ex­ tremely helpful to enable administration of chest physiotherapy to the patient with head injury. Intracranial pressure and cerebral perfusion pressure [CPP) limits are set, and when these are not ex­ ceeded. treatment is administered to ap­ propriate lung lobes or segments. Moni­ toring of ICP and CPP. therefore. may allow chest physiotherapy to be per­ formed; without it there could be no cer­ tainty that it was not detrimental to ce­ rebral perfusion. Postural drai nage and routine nursing procedures. such as turn­ ing and suctioning. may cause marked increases in ICP [See pp. 363-366). Once noxious stimulation has ceased. ICP re­ turns to baseline. provided the patient has a high cerebral compliance. There­ fore. after turning a patient. the therapist should wait a few minutes for the ICP to decrease spontaneously. rather than im­ mediately returning the patient to the baseline position if increased pressures occur. Once the patient is turned and po­ s itioned. the head may need to be propped on a roll or intravenous bag to avoid pressure on the ICP monitoring de­ vice and to decrease ICP. Percussion and vibration do not increase ICP [see Chap-

104

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

ter 8 and Appendix for details of patients with head injury treated with chest physiotherapy).

extensive physiotherapy is initiated. and this may reduce spasticity.

Turning the Patient with Abnormal Muscle Tone

Turning Patients with a Craniotomy

Decerebration or decortication may make turning the patient with head in­ jury difficult. Change of body position to minimize abnormal muscle tone allows both increased joint range of motion and improved positioning for chest physio­ therapy (Bobath. 1974). Spasticity is often affected by the position of the neck. Al­ though full cervical flexion may be lim­ ited for example. b y a tracheostomy tube. even a small amount of flexion may min­ imize extensor spasticity. The uppermost shoulder. hip. and knee may then be flexed and restrained. Flexing the hip and knee may allow patients with in­ creased lower extremity extensor muscle tone to assume the lateral position. Restraints are often required to keep the patient optimally positioned (Figs. 3.6 and 3 . 1 2 ). Once the patient with a head injury is breathing spontaneously. more

Figure 3.1 2. A sheet tied around the thigh and a wrist restraint (arrows) allow this patient to be m aintained lying sideway s. Neufeld traction is used to permit turning.

After a craniotomy. patients require proper head positioning to minimize pressure on the operative side. especially if a bone flap was removed. Placement of 500-ml intravenous infusion bags above and below this area support the head and prevent undue pressure (Fig. 3.13). Obtaining the Head-Down Position

After turning the patient. the therapist should wait a few minutes for the ICP to return to resting level. If this is below the set limit (usually 25 torr). the patient is lowered into the head-down position while the ICP and CPP are observed on the bedside monitor. Often the ICP is greater during positioning than when the patient is relaxed in the head-down po­ sition. CPP is usually> 50 mm Hg with these transient increases in ICP. If the ICP or CPP exceeds the accepta­ ble limits elevation of the head on a roll may result in a decreased ICP and in­ creased CPP. The thorax may then be lowered. Once the secretions reach the upper airway. they may be suctioned even with the head elevated. If increased ICP is a major problem. patients can be sedated before treatment in an attempt to keep the pressure as low as possible. Bar­ biturates and lidocaine may be used (see Chapter 8).

Figure 3.13. Following craniotomy the pa­ tient's head may be supported on two intrave­ nous infusion bags (arrows) to prevent pres­ sure on the operative site when the patient is turned or placed i n a head-down position.

POSTURAL DRAINAGE, POSITIONING, AND BREATHING EXERCISES

Case History 3.1. An 18-year-old male was admitted following a motor vehicle accident. He was unconscious and had a parietotemporal and basilar skull fracture, fractured mandible, multiple facial lacerations, a ruptured spleen and a lacerated liver. An ICP monitor was inserted on admission. His chest X-ray remained clear throughout the first week of hospitalization. Because the intra­ cranial pressures were greater than 15 torr when sitting up, and greater than 25 torr with turning and positioning, chest physiotherapy was not administered for 48 hr. Arterial blood gases at 8 A.M. on 60% FlO, by mechanical ven­ tilation with 10 cm H,O of PEEP were as fol­ lows: PaO" 73: pH, 7.45: and PaCO" 24. Total lung/thorax compliance (C,) before therapy was 32 ml/cm H,O. It was decided that aggres­ sive chest physiotherapy should be given be­ cause of deteriorating clinical signs, including rhonchi and decreased air entry over the right middle and lower lobes. The chest x-ray re­ mained clear, although arterial oxygenation had worsened. One hour of chest physiother­ apy was given to the right middle and lower lobes, during which ICP varied between 25 and 35 torr. Treatment produced copious amounts of viscid secretions. Air entry improved over the right base and right middle lobe: rhonchi di­ minished. Repeat arterial blood gases with the same ventilator settings were as follows: PaO" 304: pH, 7.65: PaCO" 24. Inspired oxygen was decreased to 43%, and shorter daily routine treatments were given, provided the patient was productive of secretions. C, at 2 P.M. fol­ lowing chest physiotherapy was 46 ml/cm H,O. This case history demonstrates that marked decreases in arterial oxygenation may occur and C, may worsen due to retained lung secre­ tions despite repeatedly clear chest x-rays. Ox­ ygenation and C, may be increased with chest physiotherapy. ORTHOPEDIC INJURIES

105

ible traction system that allows patients to be turned equally to the right and left. Appropriate traction is still maintained at the fracture site (Fig. 3.1 4). Neufeld trac­ tion therefore enables nearl y all the cor­ rect postural drainage positions to be ob­ tained. The patient may be moved out of bed and seated i n a chair. Ambulation training, usually non-weight-bearing, can also begin. Since turning is restricted with bilateral Neufeld traction (Browner et aI. , 1981), on admission. patients with bilateral closed femur fractures require internal or external fixation of at least one of the fractures. Tibial fractures, some humeral and pelvic fractures, and an occasional open femur fracture may be managed with exoskeletal fixation. With the exception of the pelvic fixater these devices allow turning i n all positions (Fig. 3.1 5 ). Pelvic fixation permits turning 90" to either side (Fig. 3.1 6). Alternately, a turning fr�me may be used for severe pelvic fractures with or without external fixation and al­ lows chest physiotherapy (see Case His­ tory 8 . 1 ) (Fig. 3.1 7). External fixation of fractures allows frequent dressing changes, whirlpool cleansing, and de­ bridement in patients with massive soft­ tissue injuries. It also allows patients with pelvic, femoral, tibial, and fibular fractures to begin early sitting and am­ bulation training (Hoffmann, 1954: Fel­ lander, 1 963: Karlstrom and Olerud, 1975: Edwards et aI., 1979: Brumback et aI., 1986; Burgess and Mandelbaum, 1987). Patients with a fractured acetabu­ lum or h i p dislocation are commonly placed i n Bucks traction (Shands and

Positioning the Patient with Long-Bone Injury

Some patients may be difficult to posi­ tion because of traction and splinting de­ vices or specific injuries. The therapist must be familiar with the several differ­ ent types of fixation before turning a pa­ tient with these devices. Devices that allow full patient turning and mobiliza­ tion are preferred. Closed femoral fractures that cannot be fixed by interal fixation within the first several hours of injury may be placed in Neufeld traction. This is a nex-

Figure 3.14. Neufeld traction (arrow) allows the patient with a fractured femur to be turned 90° to either side.

106

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Exoskeletal fixation for a fractured left femur and right tibia allows prone positioning.

Figure 3.15.

Raney, 1967). This form of skin traction need not i nterfere with patient position­ ing for postural drainage when there is close communication between the ortho­ pedic and therapy staff. For acetabular fractures, a small roll may be placed under the iliac crest, and a second roll below the greater trochanter, to avoid pressure on the acetabulum. With severe fractures the patient may be turned prone by turning over the opposite hip to avoid pressure on the involved hip. The Bucks traction is adjusted during turning to pull parallel to the patient's hip. Splints applied for preserving range of motion or immobilization of minor frac­ tures do not interfere with patient po­ sitioning. Extremities with fractures treated with plaster casts are easily ma­ neuvered once the plaster has dried. Soft-tissue injury alone does not inter­ fere with positioning the multiple­ trauma patient. Dislocations, for exam­ ple, are commonly seen following trauma. With posterior hip dislocation, Figure 3.16. External fixation al­ lows patients with severe pelvic fractures to be turned 90° to either side.

hip flexion and adduction are avoided. Anterior shoulder dislocations require minimizing shoulder external rotation and abduction. Patients with subluxa­ tion, dislocation, and ligamentous injury are therefore moved with caution to avoid motion similar to that causing the original injury. The following case study demonstrates the need for flexible orthopedic traction devices that allow the patient to be mobilized. Case History 3.2. A 24-year-old male was admitted following an auto accident in which he sustained a fractured first left rib, liver lacera­ tions, a retroperitoneal hematoma, an oblique fracture of the left acetabulum, and a commi­ nuted subtrochanteric fracture. Initially, for management of the pelvic and proximal femo­ ral fractures, the p atient was placed in 90-90 traction (hip flexion, 90°; knee flexion, 90°). Maximum turning was 70° to the right. and it was impossible to turn the patient onto the left Side. The admission chest x-ray was clear. Four days later, repeat chest x-ray showed

POSTURAL DRAINAGE, POSITIONING, AND BREATHING EXERCISES

Figure 3,17, This patient following thoracic spine and pelvic injuries was managed on a turning frame to facilitate skin care and chest physiotherapy treatment.

107

right upper lobe atelectasis that persisted throughout the following day (Fig. 3.1 BA). On the sixth morning the patient also developed a left lung infiltrate (Fig. 3.1 BB). At this time it was decided that a more flexible traction system was necessary to assist management of the patient's pulmonary pathology. The patient was started on antibiotics, and the traction was lowered to allow turning (Fig. 3.1 BG). Chest physiotherapy was given to the right upper and lower lobes as well as the left lower lobe. Co­ pious amounts of viscid brownish secretions were suctioned from the patient's tracheos­ tomy tube. There was clearing of the radiolog­ ical lung infiltrates (Fig. 3.1BD). This case history demonstrates that, in the trauma patient, priorities need to be estab­ lished in patient management. The traction that allowed turning was not ideal, but the patient's

Figure 3.18. (A) Chest x-ray 4 days after admission shows a right upper lobe atelectasis. (B) On the sixth day after admission a left lung infiltrate had developed in addition to the persistent right lung infiltrate. (C) The traction was lowered to permit turning and chest physiotherapy. (0) Following chest physiotherapy treatment, the lung infiltrates have cleared.

108

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

pulmonary status improved and the patient survived.

One inflexible traction system is the Thomas splint and Steinmann pin. This was traditionally used to manage femoral shaft fractures prior to the routine use of internal and external fixation devices. The pin provided a point for skeletal traction. This skeletal t raction makes turning a patient greater than 45' to ei­ ther side almost impossible (Fig. 3 . 1 9 ). If turning is restricted to 45', this prevents bronchial and segmental drainage of the posterior segment of the right upper lobe, the apical posterior segment of the left upper lobe, and the superior and poste­ rior segments of both lower lobes. This is a serious restriction as, in OUf experi­ ence, the posterior segments of the lower lobes are the most commonly atelectatic lung segments (Appendix). More flexible traction systems are preferred for pa­ tients with femoral fractures and multi­ ple trauma.

Positioning the Patient with Spinal Fracture

Patients with cervical fractures requir­ ing traction can be managed routinely with a special board that allows side to

side turning and upright positioning (Fig. 3.20). A turning frame may be used prior to surgical spinal stabilization for pa­ tients with major neurological deficit (see p. 268). Patients who cannot be turned fully on an orthodox bed can assume the prone and supine head-up and head­ down positions on a turning frame (Table 3 . 2 ). Halo vests and body casts are used for stable fractures or after surgical fixa­ tion. Halo vests are preferred to body casts because they do not restrict costal or diaphragmatic excursion, radiological interpretation, silting to 90', or skin in­ spection. Chest physiotherapy can be ad­ equately performed on patients with a Halo vest by opening the jacket. Percus­ sion and vibration may be given to the nondependent lung in the lateral position or to both lungs posteriorly when the pa­ tient is in the prone position. The vest should be opened only after the patient is appropriately positioned. Taping the opened vest to the bedrails improves tho­ racic expansion and permits percussion and vibration to most lung segments (Fig. 3 . 2 1 ). Patients with poly trauma and com­ plete neurological deficit may develop pulmonary problems if placed in a vest too early. R espiratory complications may occur due to inadequate chest expansion as noted in the following case history.

Figure. 3: 19. This patient with multiple injuries, including a pneumothorax, extensive liver and bowel Inlunes, and a fractured nght femur and left tibial plateau, can only be turned 45° to either Side due to the Thomas splint (arrow), with traction applied to the femur by a Steinmann pin.

POSTURAL DRAINAGE, POSITIONING, AND BREATHING EXERCISES

109

Figure 3.20. This quadriplegic pa­ tient following subluxation and an open-book pelvic fracture was managed in a regular bed and turned 90° to either side prior to re­ ceiving a halo vest. (Note special board and traction (arrow), which allow side-to-side turning and up­ right positioning.)

Case History 3,3, After being hit by a truck, a 45-year-old male was admitted with the fol­ lowing diagnoses: C3-C7 spinous process fractures with neurological deficit at C4, Le­ Forte I fracture, and chest injury with seventh to ninth right rib fractures. The patient was placed on a turning frame for 8 days, and he received chest physiotherapy; his chest x-ray remained clear (Fig. 3.22A). On the eighth day after admission, it was decided to place the pa­ tient in a halo vest. His pulmonary status had stabilized, although he continued to need me­ chanical ventilation due to phrenic nerve paral­ ysis. Chest x-ray the following morning showed infiltrates in both lower lobes, which became progressively worse (Fig. 3.22B). Three days after the patient was placed in a halo vest, he was returned to the turning frame and contin­ ued on aggressive chest physiotherapy (which may not have been adequate while in the halo vest). Repeat chest x-ray the following morning showed some improvement, which continued throughout the following day (Fig. 3.22C). This

patient remained on the turning frame for sev­ eral months due to his respiratory dependency secondary to lack of diaphragmatic function. The patient was later transferred to a rehabili­ tation center.

BEDS Modifications for Postural Drainage

The high incidence of respiratory com­ plications in the leu makes pulmonary care a major patient management prior­ ity. The type of bed has a significant im­ pact on the ability to perform adequate chest physiotherapy. Beds that achieve a ' 30 or greater head-down postural drain­ age position are preferred (Fig. 3.23). Gas­ kell and Webber (1973) advocate an 18inch elevation of the foot of the bed for drainage of the anterior, lateral, posteFigure 3.21. The halo jacket can be easily opened to permit percus­ sion and vibration over appropriate lung segments. Skin integrity can also be evaluated.

110

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Figure 3.22. (A) Clear chest x-ray following B days of chest physiotherapy given to the pati ent on a lurning frame. (8) Bilateral lower lobe infil­ trates developed within 24 hr of placement in the halo jacket. (e) The lower lobe infiltrates have cleared since the patient was placed back on the turning frame and received vigorous chest physiotherapy.

rior, and medial segments of the lower lobes, and an elevation of 14 inches for Ihe right middle lobe and lingula. The length of the bed and patient's position should be taken inlo consideration be­ cause they alter the tilt of the bronchial tree. Most standard beds can be placed in a 1 5· head-down position as a unit. Beds in which the head can be lowered to ob­ tain an additional tilt of Ihe bronchial tree are preferred. Shock blocks may be used with beds that do not provide an ad­ equate head-down tilt (Fig. 3.24). During the past 10 years there has been a marked increase in the usage of specialty beds with minimal evidence to substantiate their usage. Frequently used specialty beds include turning frames, low air loss beds. air fluidized beds. kinelic turn-

ing beds, and beds specifically designed for the obese patient. The need for spe­ cialty beds is primarily dependent on the nurse to patient ratio, training of the nursing staff regarding turning and posi­ tioning difficult patients, patient skin in­ tegrity, and size of the patient. Turning frames and kinetic beds are often rec­ ommended for the spinal injury patient requiring traction, although a standard bed can be adapted to provide cervical traction that allows the patient to be turned 90· to either side and sit up in bed for pulmonary care (see Fig. 3.20). With this system spinal stabilization is ade­ q uate for management of cervical flexion injuries (Frederick Geisler MD, personal communication). It is the author's opin­ ion that the turning frame allows beller

POSTURAL ORAINAGE, POSITIONING, AND BREATHING EXERCISES

Hydraulic beds are easily placed in the head-down position.

Figure 3,23.

This electric bed is placed on shock blocks to promote more adequate pos­ tural drainage.

Figure 3.24.

111

skin inspection, chest physiotherapy treatment, and early rehabilitation than a kinetic bed. Proponents of all specialty beds claim improved skin condition com­ pared to a standard bed, although studies evaluating turning frequency and body position on standard versus specialty beds are minimal. I t has been the au­ thor's personal observation that low air loss beds and flotation beds are often util­ ized to replace side to side turning. Turn­ ing affects all body systems, not just skin pressure. Air fluidized beds offer contact pressures lower than capillary closing pressures. Decreased capillary pressure is thought to lower the incidence of tis­ sue breakdown and aid healing. Hargest [1977) states that pneumonia has not de­ veloped in patients on air fluidized beds whose primary problem is lack of move­ ment. lt is our experience and that of oth­ ers [Smoot, 1986) that retained secretions and pneumonia do develop in trauma pa­ tients while on an air fluidized bed. Sit­ ting upright and performing postural drainage are extremely difficult on this bed [Fig. 3.25). The continuous movement of kinetic beds is thought to minimize tissue break­ down and secretion retention. Schimmel et al. [1977) demonstrated use of the ki­ netic bed to change ventilation-perfu­ sion relationships. In a patient with a gunshot wound to the chest, a right lung contusion cleared, but a left lower lobe atelectasis was apparent on the chest x­ ray taken within 8 hr of admission. At­ tempts to perform chest physiotherapy on patients i n this bed demonstrate that adequate treatment can be given only for four lung segments: the anterior seg­ ments of both upper lobes, the right mid­ dle lobe, lingula, and anterior segments of both lower lobes. Seven postural drainage positions for the other lung seg­ ments therefore cannot be obtained. as the bed limits turning. In practice. the pa­ tient cannot be positioned with the af­ fected lung uppermost for the most com­ monly atelectatic lung segments. The manufacturer advocates removal of the posterior chest portion of the bed to allow chest physiotherapy treatment. However. percussion and vibration without the use of segmental postural drainage is likely to be ineffective [see Chapter 4). Patients

112

CHEST PHYSIOTHERAPY I N THE INTENSIVE CARE UNIT

Figure 3.25. The height and depth of the low air loss bed make mobi­ lizing patients in and out of bed very difficult. Five staff members are required to lift this mechanically ventilated multiple trauma patient from the low air loss bed to a chair.

who do not tolerate positioning and turn­ ing on a turning frame because of in­ creased ICP or agitation also do not tol­ erate treatment on the kinetic bed. For patients with increased ICP. a 45' head­ elevated position is difficult to achieve. Table 3.2 compares standard and spe­ cialty beds. The following case history demonstrates the disadvantages of the ki­ netic bed for a patient with severe chest trauma. Case History 3.4. A 1 9-year-old female was admitted to the trauma center following a mo­ torcycle accident. She had a right pneumotho­ rax, lung contusion, rib and clavicular frac­ tures, and a torn right main stem bronchus. She was taken to the operating room for rean­ astomosis of the torn bronchus and a laparot­ omy. liver lacerations were found and repaired. After surgery the patient was placed on a ki­ netic bed. Repeat chest x-ray 8 hrs after ad­ mission showed the bilateral alveolar-intersti­ tial infiltrates of respiratory distress syndrome (Fig. 3.26). This pattern persisted during the patient's hospital stay. The kinetic bed was used according to manufacturer's instructions when the patient was stable hemodynamically. No chest physiotherapy was given because of inability to achieve proper bronchial drainage positions. Chest x-ray 4 days after admission showed complete right lung atelectasis (Fig. 3.27). The patient's PaO, was 45 on an FlO, of 1 .00. Bronchoscopy was performed; copious amounts of retained secretions were noted. The anastomosis in the right main stem bron­ chus remained intact. Cardiopulmonary arrest developed following bronchoscopy secondary to hypoxemia. The patient expired after numer­ ous attempts at resuscitation, including open cardiac massage.

The kinetic bed was unable to prevent the development of complete right lung atelectasis in this patient with chest trauma. Chest phys­ iotherapy performed with the patient turned into the postural drainage positions described in this chapter may have prevented some of the aspiration of blood and secretion retention. Once complete atelectasis and hypoxemia de­ veloped, they were irreversible.

This bed did not result in a favorable outcome with this patient. It may be ad­ vantageous for quadriplegic patients in a chronic facility, who have recurrent chest infections and decubitus ulcers. Often at home or in these facilities there is insufficient help for adequate patient turning. Bedridden quadriplegic patients who experience severe shoulder pain with side-to-side turning may prefer the kinetic bed. Results similar to the man­ ufacturer's claims of treating acute atel­ ectasis are obtained with chest physio­ therapy (see Case History 6.1 ). When working in a critical care unit it must be realized that any time a thera­ peutic intervention interferes with the operation of a specialty bed the cited benefits are eliminated. Many patients require bedside medical therapy, special studies, and transportation from the ICU for special procedures. Proponents of the rotating bed claim that the hazards of im­ mobility are minimized and eliminated, although the bed is often immobile (Trammel et aI., 1985), particularly when used for critically injured trauma pa­ tients (see Appendix, pp. 361-362). Initial claims by the manufacturer that kinetic therapy eliminates the hazards of immo-

Table 3.2 Comparison 01 Standard and Specialty Beds Standard Bed

Criteria Recommended patient population

Patient evaluation and inspection Bedside diagnostic tests

Turning ease and frequency

Patient comfort

All

Large Person

Turning Frame

Low Air Loss

When body weight exceeds standard bed weight requirements, 700 tb limit

Spinal injury or patients requiring inspection and treatment posteriorly. 250 Ib limit

Immobilized patients, except spinal injury, tissue breakdown, burns

Same

Same

All body surfaces are Same exposed with side to side turning Same. 90° upright Allows 90° upright easier to obtain and lateral positioning for bedside x·rays and special procedures Dictated by the Same. 3-4 persons may patient's medical status, usually be required every 2-3 hrs, 1 -2 persons for routine turning. 3 for obese patients or prone positioning Subjective, normal environmental stimulation

Subjective

Cannot obtain true Same upright position

Manufacturers suggest 1 person, in leu 2 persons every 2 hrs, prone and supine poSitions

Subjective, some patients do not like prone position

Same

Subjective

Air Fluidized

Kinetic

Burns, severe skin lesions. Spinal injury requiring traction, immobilized immobilized patient up 10 260 lb. Nol patients, skin pressure recommended with sores, questionable for cardiac disease, patients with elevated ICP. reduced lung function, Patient must be able to tolerate rotation 1 8 hr/day or disoriented patients (Kalaja. ( 984) Same Only anterior body surface is exposed. Pad placement restricts inspection Cannot obtain true upright Cannot obtain true upright; position. difficult for abdominal and chest films are more difficult to obtain placing cassettes to take bedside x·rays. and interpret. Cassette rack more difficult to interpret placement is not ideal due x·rays to varying patient size Manufacturers claim these 1 24° every 3.5 min while rotating. One person beds eliminate the need for patient turning required to operate bed although turning is necessary for most body systems (see Chapter 6); nursing staff often neglects necessary turning (Smoot. ( 986) Subjective, disorientation Subjective, manufacturers and has been reported Keane (1977) claim better (Smoot. 1 986; Lucke than the Stryker frame; 4 and Jarlsberg, 1 985; stroke patients found Rath and Berger 1 982). confinement intolerable Patients have requested (Kelley el al.. ( 987); to be removed from bed increased agitation has (Bolyard el aI., 1987) been noted in head injured and complained of patients; 30% patients on weightlessness, inability bed requested it be to move freely. and stopped (Trammell et aL, elevate head of bed ( 985) (Nirmille and Storm ( 984)

'U 0 (J) -< C JJ l> r 0

JJ

� Z »

Q !" 'U 0 (J)

=i 0 Z Z

.P » Z

0 III

JJ

m

»

-< I

Z

Q m x

m JJ

()

Ui m

(J)

� � ...

� �

...

Table 3.2 (Continued) Comparison of Standard and Specialty Beds Criteria Pulmonary care

Spinal alignment

Intracranial pressure

Standard Bed

Large Person

Turning Frame

Low Air Loss

Suctioning: no interference

Suctioning: same

Suctioning: same, prone position assists oropharyngeal drainage

Suctioning: same

Postural drainage: all postural drainage positions

Postural drainage: 5/11 drainage positions obtained. Does not go into head·down position

Postural drainage: 7 of 1 1 postural drainage positions, prone position assists drainage of posterior and most frequently involved lung segments

Postural drainage: same

Maintained through standard traction or special board, (see Fig. 3.20) Head may be raised as necessary to lower ICP

Same

Same

Adequate for most spinal injuries

Head can be elevated to reduce ICP

Air Fluidized

Kinetic

Suctioning: manufacturers claim the body position changes improve cannulation of the left mainstem bronchus. This is not substantiated in the literature (Kirimli at aI., 1970; Kubola el al.. 1 980) Postural drainage: allows Postural drainage: More 1 2° head-down position; 4/ difficult to obtain the 1 1 1 1 positions can be positions; head·down and obtained. Results same as sitting positions are not documented by Mackenzie optimal. Atelectasis and el aI., (1 985). Pulmonary decreased lung function complications may be documented (Smoot, 1 987: reduced in spinal injury Kalaja, 1 984) patients not easily turned (Reines and Harris, 1 987) Decreased atelectasis, pneumonia, in trauma patients although frequency of turning on a conventional bed is not documented (Genlilello el al., 1 988) Suctioning: same

Not recommended for spinal stabilization

Not recommended for spinal stabilization

Same

Difficult to maintain upright position; when patient is upright, posterior thorax does not receive benefit of bed

Proponents claim better than turning frame, not documented by controlled studies Kelley el 01., (1 987) found increased transtentorial herniation compared to standard bed, ICP not significantly altered by stationary bed positions (Gonzalez·Arias et aI., 1 983). Affect while rotating unkown

() I

m (f) -i

" I -
1 00 mm Hg

30 adults. following open heart surgery

20 newborns

Design

Hyperoxygenation

similarly prevented hypoxemia: hypoxemia noted in both groups 3. Hypoxemia was equally prevented by bolh melhods; no added benefit from O2 insufflation; preoxygenation/ hyperinflation equally effective as using the port adaptor

No difference in Pa02 using other melhod; pH I with bagging but T with sighing O2 insufflation resulted in less

=

=

() ::r

m en

-i

."

::r -
-< () r m :> :D :> z ()

r.'

()

0

C Gl

I Z

Gl

:> z

Hyperinflation Preoxygenation Hyperoxygenation

Preoxygenation Hyperoxygenation Hyperinflation

Bagging with PEEP caused a significant T in Pa02

Used intermittent suctioning

Only procedure 3 had significant 1 in Pa02 during suctioning Procedure 4 was associated with significantly less 1 Pa02 and 1 and 2 during suctioning All procedures after

Hyperinflations 150% of baseline VI FI02 was 1 to 1 .0 for 1 min pre- and/or postsuctioning Hyperinflation should be aborted if hypotension or bradycardia occurs Recommend 1 min of 1 00% O2 before and

0 U> C ()

-i 6 Z

Z Gl =

�

....

Co>

� ..... ...

Table 5.5 (Continued) Recent Studies on Preventing Hypoxemia during Suctioning ... Investigator

Brown at al (1983)

Subjects

22 acutely ill patients, most with preexisting COPD

Design before and after suctioning 3. Hyperinflation before and after suctioning 4. t FlO, and hyperinflation before and after suctioning Phase I: compared no extra breaths to 4 prescutioning breaths, 4 postsuctioning breaths. and suctioning through an adaptor (all at baseline FI02l Phase II: compared suctioning using an adaptor to 6 presuctioning breaths and! or 6 postsuctioning breaths at FlO, 1 .0 Phase III: compared 4 successive catheter passes using -

Adjuncts Used

Conclusions suctioning were successful in restoring Pa02 All procedures were equally effective at 5 and 1 0 min after suctioning

Preoxygenation Hyperoxygenation Hyperventilation

Phase I: greatest desaturation occurred with no hyperventilation; desaturation significantly less with adaptors; recovery best using the adaptor or 4 postsuctioning breaths Phase II: significantly more desaturation occurred with only 6 postsuctioning breaths; adaptor equally effective to other two methods Phase III: adaptor equally effective at preventing desaturation

Comments atter suctioning

Used intermittent suctioning FI02 t 1 min before extra breaths initiated Recommend using the adaptor without altering ventilator settings

o I m (f) -i " I -< (f)

� I m JJ > " -
30 ml ) sputum and the cardiorespiratory changes that are shown in Table 7.6. Despite the production of considerable quantities of sputum in both patients only the 48·year-old patient with previously normal lungs showed beneficial effects from sputum removal (Table 7.5). QJQ" Q" pulse, and PaO, all fell in this patient but remained unchanged in the other (Table 7.6).

Cardiac and Respiratory Function

Both CT and R.w values shown in Fig­ ures 7.4 and 7.5 suggested that following chest physiotherapy, secretions were re­ moved from the small airways. To inves­ tigate this possibility further and to quantitate the changes taking place with objective measurement, cardiac and res­ piratory function were measured before and after chest physiotherapy. Nineteen

trauma patients were studied (Macken­ zie et al.. 1985). Before investigation, all were mechanically ventilated for at least 6 hr with PEEP (Mean, 9 em H,O, range 5-18) and had indwelling arterial and thermistor-tipped, pulmonary arterial lines in position. Intrapulmonary shunt (0,/0,), dead space (V./V,), cardiac out­ put (0,), and CT were measured before, immediately after, and 2 hr after chest physiotherapy. The indications for ther­ apy included secretion retention with segmental or platelike atelectasis (11 patients), lung contusion (6 patients), and respiratory distress syndrome (2 patients). Respiratory failure following trauma may frequently result in high 0./0, that requires increasing levels of inspired ox­ ygen (FlO,) and PEEP to maintain ade­ quate oxygenation. If chest physiother­ apy in the presence of retained secretions can reduce the requirements for high FlO, and PEEP, it is beneficial. It may also result in reduced morbidity and mortal­ ity from respiratory failure after trauma.

Table 7.5 Cardiorespiratory Changes in a 48-Year-Old Male Patient after 60 Min of Chest Physiotherapy Variable' FlO, PaO, (torr) PaCO, (torr) a-vDO,(ml) Pulse (beats/min) P� (torr) PCWI' (torr) Q, (I/min) R.O', (dynes/sec/em ') R", (dynes/sec/em ') QJQ, (%) C, (ml/cm H,O) YO, (ml/min)

Before Physiotherapy 0.61

U

48 8.2 125 68 25 8.7 405 1 29 29.4 17 656

Immediately after Physiotherapy

Two Hours after Physiotherapy

0.61 71 32 7.7 1 10 80 20 6.0 800 1 33 1 7.3 20 460

0.61 1� 33 3.6 1 00 95 24 5.4 1 051 1 33 20.7 26 194

'a-vD02, arterial venous oxygen difference; Part. mean arterial pressure; R, resistance; V02• oxygen consumption per minute.

226

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Table 7.6 Cardiorespiratory Changes in a 54-Year-Old Male Patient after 50 min of Chest Physiotherapy Variable

Before Physiotherapy

Immediately after Physiotherapy

Two Hours after Physiotherapy

FlO, PaO, (torr) PaCO, (torr) a-vOO, (ml) Pulse (beats/min) p,. (torr) Pew. (torr) a, (liters/min) R,,., (dynes/see/em-') R"", (dynes/see/em-') 0./0, (%) C, (ml/em H,O) VO, (ml/min)

0.54 1 12 37 5.0 1 25 95 21 7.9 61 8 1 32 1 7.5 48 394

0.54 1 24 39 5.3 122 90 21 7.6 745 1 36 14.8 36 403

0.54 95 37 37 125 1 00 18 7.7 858 165 1 8.7 41 368

a-vD02• arterial venous oxygen difference; p.rh mean arterial pressure; R. resistance; 'iJ02• oxygen consumption per minute.

Chest physiotherapy produced a signifi­ cant fall in 0./0,. The greatest decrease in 0./0, was from 36.3% to 16.5% a nd 13 patients showed a fall in 0./0, immedi­ ately after therap¥. In six patients there was a rise in O./Q, (greatest from 14.8% to 23.5%) that persisted for at least 2 hr in three of the patients with lung contusion (Fig. 7.11). Dead space, PaCO" and PaO,

were unchanged (Table 7.7). There was a significant overall increase in total CT 2 hr after CPT. Eleven patients showed a rise in CT immediately after and 2 hr after CPT. In three patients CT was unchanged and in five CT fell. The greatest CT rise was from 34 to 51 mljcm H20, while the greatest CT decrease was from 48 to 36 mljcm H,O. In three patients, all treated 40

Figure 7.11. Individual plots of 0./0, before, immediately after, and 2 hr after a single chest physiotherapy treatment. (From C.F. Macken­ zie et al.: Crit Care Med 1 3:483-486.1985.)

30

....0

0 0

20

10

:

' -�

.

I�

-= •

.------ . 0 PRE

POST

2HRS POST

227

PHYSIOLOGICAL CHANGES FOLLOWING CHEST PHYSIOTHERAPY

Table 7.7 Cardiorespiratory Function before, Immediately alter, and 2 hr alter CPT Before CPT CI (Iiters/min/m') LVSWI (g/m/m') RVSWI (g/m/m') Ca-V O, (ml/dl) Pulse (beats/min) 0,/0, (%) V,jV, C, (ml/cm H,O) PaO, (torr) PaCO, (torr)

4.5 ± 65 ± 19 ± 4.2 ± 94 ± 16.4 ± 0.47 ± 29 ± 128 ± 34 ±

Im mediately after CPT

1.35' 27.1 6.0 1.36 23.2 7.55 0.16 11.3 36.6 7 .4

4.2 51 17 4.2 95 13.2 0.46 32 143 35

± ± ± ± ± ± ± ± ± ±

1.23 27.7 6.6 1.38 23.0 4.03" 0.15 13.4 31.5 8.1

Two Hours after CPT

4.0 63 18 3.9 90 14.5 0.45 33 137 33

± ± ± ± ± ±

0.92 27.3 6.6 0.91 23.5 6.10 ± 0.13 ± 11.r ± 24.9 ± '6.7

From C.F. Mackenzie et al.: Crit Care Med 13:483-486,1985.' Mean ± SO." P < 0.05 before and immediately after."" p < 0.05 before and 2 hr after.

for atelectasis, the fall in CT persisted for hr (Fig. 7.12). The mean values of car­ diorespiratory function for the 11 pa­ tients with atelectasis were not clinically or statistically different from those of the six patients with lung contusion. There were no hemodynamically significant cardiac dysrhythmias during or 2 hr after CPT. There was no significant change in car­ diac function after CPT (Table 7.7). But no measurements were made during therapy because positional changes alter 2

70

Figure 7.12. Individual plots of CT before, im­ mediately after, and 2 hr after a single chest physiotherapy treatment. (From C.F. Macken­ zie et al.: Crit Care Med 13:483-486.1985.)

65 60 0N 50 :I: E

" ...

E

� u

40 30 20 10

cardiac function (Rivara et aI., 1984) and cause ventilation perfusion changes (Ka­ neko et aI., 1966; Douglas et a!.. 1977; Zack et al.. 1979; Remolina et aI., 1981). Decreases in mixed venous oxygenation occur during suctioning (Bade et aI., 1982) and together with change in posi­ tion may explain the lack of significant change in PaD, despite a fall in G./G, im­ mediately after CPT. There are three other reports of cardiac function after CPT in the literature (Laws and Mc­ Intyre, 1969; Barrell and Abbas, 1978

./� '� L.' /

�"-

-/ -'

:.� /�� �

PRE

POST

2 HRS POST

228

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Klein et ai, 1988). All reported detrimen­ tal effects. In the Laws and Mcintyre study cardiac output (by dye dilution) in­ creased by 50% with artificial coughs produced by lung hyperinflation. Barrell and Abbas found mixed venous oxygen­ ation and cardiac output fell significantly during CPT but returned to baseline val­ ues within 10 min after the end of ther­ apy. The differences may be accounted for by CPT techniques and patient pop­ ulation studied. Laws and Mcintyre 's pa­ tients had respiratory failure and re­ ceived inflation pressures during the artificial cough of 60-100 cm H,O, which many patients found distressing. Barrell and Abbas studied 14 patients who were extubated after mitral valve replace­ ments. Despite the changes in cardiac output and mixed venous saturation, ar­ terial oxygenation was unchanged. The same effects were not found in our study of 19 trauma patients. This difference may be because the trauma patients were young and did not have preexisting car­ diac disease. In addition, falls in mixed venous saturation may only be transient and occur during CPT. All the trauma pa­ tients were ventilated with PEEP, which is known to restore functional residual capacity and improve oxygenation (Mc­ Intyre et aI., 1969). The increase in CT in 11 patients was probably secondary to the recruitment of more functioning al­ veolar units as a result of mobilization and clearance of secretions from small airways. The process of recruitment may have been assisted by PEEP and interdependence. Klein et al. (1988) showed that cardiac output did rise by 50% over baseline val­ ues during chest physiotherapy in two unsedated patients. In patients who re­ ceived continuous infusion of 3 I'g/kg fentanyl there was still a 20-25% rise in cardiac output during chest physiother­ apy that returned to baseline within 15 min after the end of therapy. With the analgesia there was no significant in­ crease in heart rate or blood pressure but 0, consumption and CO, production were still increased. The important ques­ tions about changes in cardiac function occurring during chest physiotherapy are: Are they clinically relevant? Are they associated ",ith detrimental long

lasting effects? Can they be prevented with modifications of chest physiother­ apy and use of analgesia and sedation? While the decreased 0./0, in three of the contusion patients may suggest an adverse effect of CPT, contusion is a rec­ ommended indication for CPT (Macken­ zie and Shin, 1978b; Richardson et aI., 1982). The greatest decrease in 0./0, also occurred in a patient with lung contu­ sion. The rise in 0./0, may reflect trans­ bronchial aspiration of bronchial secre­ tions during positioning for CPT (see Case History 7.1). In patients with lung contusion, in whom coagulopathy com­ plicates management, CPT should be given to the noncontused lung after ther­ apy to the contused area. Alternatively, endobronchial intubation with a double lumen tube may prevent transtracheal aspiration of blood. This study confirmed our clinical impression that in critically ill patients, mechanically ventilated with PEEP, who have low lung compliance and increased 0./0" CPT does not pro­ duce the deleterious cardiopulmonary changes that have been reported with bronchoscopy (Lundgren et al.. 1982). CPT may be used to manage retained lung secretions due to acute posttrau­ matic respiratory failure, without pro­ ducing hypoxemia, Mass Spectrometry Analysis of Expired Gases

Mass spectrometry, as peviously de­ scribed (McAslan, 1976), was used to an­ alyze breath-by-breath end-tidal carbon dioxide (PHCO,) and oxygen (PHCO,) in mechanically ventilated patients, To de­ termine the effects of chest physiother­ apy on the uptake and excretion of 0, and CO" analyses were obtained before, during, and after treatment. Case History 7,3, An 18-year-old male was admitted following an automobile accident. He had a severe head injury, pulmonary edema, and had suffered cardiac arrests 8 and 10 hr previously. He was given chest physiotherapy because of a complete atelectasis of the right lower lobe and deterioration in arterial oxygen­ ation while mechanically hyperventilated, with an FLO, of O.S. The abnormal traces obtained by mass spectrometry before chest physio­ therapy are shown in Fig. 7.13. Note the ftuc­ tuations seen in expired CO, during the latter

PHYSIOLOGICAL CHANGES FOLLOWING CHEST PHYSIOTHERAPY

�2 t.

t 1

2:[

229

..COftd--t ,

r "

Figure 7.13. Breath-by-breath 0, and CO, analysis in the patient before chest physiotherapy. Note the lack of alveolar plateau and the presence of a terminal peak in the CO, curve. P.,. 0, and P"CO, values are shown. part of expiration. ABG analysis showed a PaO, of 69, a pH of 7.51. and a PaCO, of 30. After 45 min of chest physiotherapy to the right lower lobe and removal of copious secre­ tions. there was clinical and radiological evi­ dence of clearance of the atelectasis. The ex­ pired gas wave form now showed a plateau (Fig. 7.14). The trace in Figure 7 .14 was made 1.5 hr after that in Figure 7.13. ABGs sampled at the same time as the mass spectrometry analysis in Figure 7.14 showed a PaO, of 148. a pH of 7.52. and a PaCO, of 34 despite un­ changed ventilator settings and FlO,.

The baseline traces obtained before therapy show a lack of constant "alveo­ lar" CO,. suggesting that the CO, ten­ sions did not reach the same values as the pulmonary capillary CO, in all alve­ oli. Therefore. a considerable arterial al­ veolar CO, gradient (a-ADCO,) would exist. P.-rCO, levels were inconsistent; well-ventilated alveoli emptied early; poorly ventilated units in which alveolar gas was trapped emptied later. There is a peak in the terminal plateau of the CO, curve (Fig. 7.13). Similar peaks in me-

FO 0. 0.6

'h a, 395 torr P[T eo. 29

CO.

29 o

�

"'"

chanically ventilated patients may be abolished with PEEP (McAslan) or an ex­ piratory retard (Ricker and Haberman. 1976). p."CO,. in the presence of this ter­ minal increase in expired CO,. is unrep­ resentative of "average" mixed alveolar gas composition (McAslan. 1976). There­ fore. the recorded a-ADCO, of 4 torr may be only an estimation of the gradient. This situation is thought to occur due to V/0 imbalance and premature airway closure. If there is a large difference in the distribution of ventilation causing a spectrum of rates of emptying. the slope of the alveolar plateau increases. Time constants given as the product of regional lung compliance and airway resistance describe the rate of airway emptying dur­ ing expiration. In obstruction. time con­ stants increase so that no alveolar pla­ teau may be visible on the expired CO, curve (Fig. 7.15). The abolition of the peak by chest physiotherapy (Fig. 7.14) and the change in shape of the expired CO, curve appeared to indicate that chest physiotherapy affected the small airways Figure 7.14. Following chest physiotherapy. an alveolar plateau is apparent. and the terminal peak in the CO, curve is abolished. Note the rise in the P.,.CO, value after chest physiotherapy.

230

-6

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT , - - - - 2 sec -- .

S C02

-5

-4

-3

_2

_1 -0

Figure 7.15. Expired CO, curve of patient with chronic bronchitis and emphysema. Note the lack of alveolar plateau suggesting inequality of emptying of alveoli and impaired distribution of ventilation.

and their ability to take part in gas ex­ change. The expired gas wave form after chest physiotherapy shows a plateau with a constant value for P"CO,. The a­ ADCO, of 5 torr is increased. This sug­ gests that alveolar dead space is similarly increased after chest physiotherapy and removal of secretions. The expired CO, curve of a patient out of phase with a mechanical ventilator shows a characteristic change (Fig. 7.16A). This occurs because the patient attempts to breathe in non-CO,-contain­ ing gases while the ventilator is in the ex­ piratory phase. This patient was thought to be out of phase because of lung secre­ tion retention. Following chest physio­ therapy and removal of these secretions, a normal CO, curve was produced (Fig. 7.16B). Breath-by-breath analysis of CO, and 0, curves during chest physiotherapy and chest wall vibration is shown in Fig­ ure 7.17. The CO, and a, analysis con­ firm that chest wall vibration causes changes in the expired gas wave form. These variations occur due to the ex-

pired air column oscillations found with chest vibration. Mass spectrometry dem­ onstrates that chest physiotherapy pro­ duces changes in CO, and a, gas ex­ change. The more normal CO, curves found following chest physiotherapy sug­ gest that the therapy has a favorable ef­ fect on the small airways. ANALYSIS OF CH EST PHYSIOTHERAPY DATA

Factors Influencing Physiological Measurement in the leu

Physiological measurement in the ICU is difficult. The patient, especially when unstable, undergoes continual and often dramatic changes in cardiorespiratory function. Because the patient may be critically ill, therapeutic intervention is frequently necessary, and this often al­ ters cardiac or respiratory function. While it is usually quite easy to exclude such obvious cardiorespiratory changes as development of shock or tracheal in-

Figure 7.16. (Al Breath-by-breath CO2 curve in a patient who was out of phase with mechanical ventila­ tion. (8) CO, analysis on the same patient after chest physiotherapy shows a normal curve.

8 Normal Co1ant ofler chell phYliolheroPJ

\-L." "'-I

PHYSIOLOGICAL CHANGES FOLLOWING CHEST PHYSIOTHERAPY

VIBRATI ON DORING EXPIRATION--+ ,

!5�

".

�

231

I l ,

!1 �

!

r-r-

f---1S........ CO.

l·· " ,_

/' M. (

('if I

�

J\ (V' f\-

\ " " , ""\ -� �.�,..._ ....�-- � _ �-... _

\

\.

r r r"

f'1, "

"

-

,,

\,

r (' ,,

I +

�

ills: ..

....,....

Figure 7.17. Breath-by-breath 0, and CO, analysis during vibration chest physiotherapy. showing a normal curve when vibration ceased and loss of the curve with suctioning of the tracheobronchial tree.

tubation and mechanical ventilation, other more subtle variables may easily go undetected or be attributed to the ther­ apy under study. This is one argument for using a control population. Investiga­ tion of chest physiotherapy in ICU pa­ tients is made especially difficult because there is no control or standardization of the therapy itself. Many reports of chest physiotherapy include the use of bron­ chodilating or mucolytic agents. Some centers do not use postural drainage, while others exclude chest vibration of percussion. The duration and frequency of therapy vary enormously (see p. 39). What control should be used to com­ pare with chest physiotherapy in criti­ cally ill patients? The commonly sug­ gested control is side-to-side turning and tracheal suction (Murray, 1979b). How­ ever, when researching the effects of therapy in humans, both standards of care must be comparable and acceptable medical practice. In our experience, side­ to-side turning and tracheal suctioning are not comparable therapy for treatment of conditions such as left lower lobe, pos­ terior segment atelectasis. Long-term fol­ low-up studies suggest that this regime did not prevent deterioration in chest x­ ray appearance, PaO,/FIO" and CT (see pp. 233-237). Also, when side-to-side turning and suctioning are used. as oc­ curred at our institution during a period of understaffing of physical therapists, at­ electasis was not prevented. In October 1977. there was a physical therapy staff

shortage and a large patient load. Twenty-one patients in the step-down (ICU) unit were not treated by physical therapists but. instead, received tracheal suctioning and side-to-side turning. In that month, there were 14 readmissions to the critical care recovery unit (CCRU) from the ICU because of deteriorating respiratory function that was docu­ mented by clinical examination, chest x­ ray, and blood gas analysis (c. F. Mac­ kenzie, personal communication to R. A. Cowley. M.D., Director MIEMSS, October 1977). In contrast, an average of three readmissions/month occurred during the remainder of the year. This strongly suggests that side-to-side turning and suctioning are not comparable or accept­ able medical practice, compared to chest physiotherapy, when performed on ICU patients with acute lung pathology. In addition to the ethics of using con­ trol groups in critically ill patients, and the difficulty of finding a comparable and clinically acceptable control, some stan­ dardization of chest physiotherapy should be achieved before physiological measurements can be truly claimed to re­ flect the effects of chest physiotherapy. Factors Influencing Analysis of Data from Dillerent ICUs

Therapist Variability

When comparisons are made of chest physiotherapy between different ICUs, is it possible to exclude such a subjective

232

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

but very important factor as the clinical variability between one therapist's abil­ ity to clear an acute atelectasis and an­ other 's inability? It is our impression that physical therapists trained by staff expe­ rienced in ICU work are more successful in improving the patient's clinical and ra­ diological signs than are nurses, respira­ tory therapists, or physicians. This im­ pression is confirmed by the following case history. Case History 7.4. A 23-year-old male was admitted with a head injury, cerebral contusion, and a ruptured spleen sustained in an auto­ mobile accident. He was tracheally intubated, mechanically hyperventilated (paCO, < 30 torr) and given corticosteroids. Five nights after admission (that is, November 1 8) he developed a complete left lower lobe atelectasis. This was recognized following deterioration of arterial oxygenation and after a chest x-ray. The pa­ tient was suctioned, FlO, was increased, and the ABG analysis was repeated. little improve­ ment occurred as is shown in Table 7 . 8). After suctioning, a nurse gave chest therapy with the patient turned onto his right side. There was an increase in arterial oxygenation from 7 1 to 1 1 8 torr and of CT from 44 to 60 ml/cm H,O. The patient was treated 3 hr later by a chest phys­ ical therapist in the correct postural drainage position, after which arterial oxygenation fur­ ther improved to 1 85 torr and CT increased to 68 ml/cm H,O. The following morning (November 20), arte­ rial oxygenation on the same ventilator settings had again deteriorated despite tracheal suc­ tioning and "chest therapy" performed by the nurse. Chest physiotherapy by a physical ther­ apist again produced improvement in arterial

oxygenation and increase in Cr> The patient, who regained consciousness, was placed on a T-piece and was successfully extubated.

Other investigators confirm that trained physical therapists produce a more favorable outcome than do nurses or physicians. Vraciu and Vraciu (1977) found that breathing exercises adminis­ tered by a physical therapist reduced the incidence of pulmonary complications after open heart surgery, compared to turning, deep breathing, and coughing every hour assisted by nursing staff. Finer and Boyd (1978) noted that the im­ proved oxygenation that occurred after chest physiotherapy in infants was re­ lated to a chest physiotherapist, rather than to the ICU nurses, performing chest physiotherapy. Lyager et al. (1979) thought that the reduction in pulmonary complications that they reported, com­ pared to those noted by Bartlett et al. (1973), may have resulted from their use of specially trained physical therapists rather than residents or nursing staff. Ap­ plication of a standard therapy, such as is advocated in this book, may help reduce therapist variability. Patient Population

How can compensation be made for the variability between different patient populations? For example, can chest physiotherapy or any other therapy be compared between trauma and medical ICUs? Admission to a trauma unit is not

Tabte 7.8 ABG, C" and PaOJFIO, Changes Occurring over a 3-Day Period

Date

Time

PaO, (torr)

pH

PaCO,

CT (ml/ cm H,O)

PaOJ FlO,

November 1 8

1 1 :00 21 :30 23:00 05:00 10:00

1 07 57 71 118 1 85

7 . 58 7.51 7.59 7.58 7.60

28 31 28 27 28

78 68 44 60 68

282 1 50 1 61 268 420

22:00

94

7.54

33

60

214

09:40

1 46

7.57

31

78

348

10:45

122

7.56

35

14:00

112

7.52

36

November 1 9

November 20

Event Left lower lobe atelectasis Suction, FI02 increased Chest therapy by nurse Chest physiotherapy by physical therapist ABG fall at night Chest physiotherapy by physical therapist ABG following chest physiotherapy Spontaneous respiration Tpiece 40% O2 Extubated face tent 0,

PHYSIOLOGICAL CHANGES FOLLOWING CHEST PHYSIOTHERAPY

determined by a history of preexisting pathology but usually by an acute insult to an otherwise-healthy individual. Ap­ plication of physiotherapy is likely to have a different effect in the trauma pa­ tient than in the medical patient. It may be possible to overcome this variability by considering a large patient population studied over a long period. Method of Mechanical Ventilation

Manufacturers of ventilators and ad­ vocates of different modes of mechanical ventilation and respiratory support claim that important differences result from use of one or other ventilator or methods of ventilation. If this is so, the outcome of chest physiotherapy in mechanically ventilated patients is likely to be differ­ ent, even when the patient population and therapy are similar. Variability in Anesthetic Techniques

Different anesthetic techniques pro­ duce different effects on the respiratory system. Regional anesthetic techniques are credited with causing fewer deaths after surgery than general anesthetic techniques (Beecher and Todd, 1954). However, there is a lack of controlled studies comparing regional anesthesia to modern general anesthetic techniques. Regional techniques are frequently not applicable, as, for example, in the man­ agement of a patient with multisystem trauma. Regional techniques such as epi­ dural anesthesia are time consuming and require considerable expertise. Nonethe­ less, they are of enormous benefit for re­ lief of pain after surgery, when an epi­ dural catheter is left in place. Regional analgesia is more effective than narcotics for maintaining pulmonary function after surgery (Fairley, 1980). This may consid­ erably alter the incidence of respiratory complications and the need for mechan­ ical ventilation (see Chapter 10, p. 341 for details of pain relief in ICU). General anesthesia with use of neuro­ muscular blockade (and an opiate, or low doses of inhalational agent) frequently produces a more awake patient, on rever­ sal of the neuromuscular blockade, than occurs following the use of inhalational agents alone. The recovery period, to the

233

point where the patient sits up, follows commands, and takes deep breaths and coughs, may be shortened when neuro­ muscular blockade is used. This may de­ crease the incidence of respiratory complications. Water vapor loss may be reduced by use of a closed or semiclosed rebreathing circuit on the anesthesia machine. Alter­ natively, inclusion of a humidifier or nebulizer in the ventilator/anesthesia machine ensures humidification of dry anesthetic gases at all times. Lack of hu­ midification during anesthesia or in the period after surgery may be an important determinant of subsequent respiratory complications in the ICU. (See Chapter 9 for more details on humidification.) Therapeutic Intervention

In the long-term follow-up of patients in ICUs, can the numerous other clini­ cally essential therapeutic interventions (such as intravenous fluids, vasoactive drugs, and analgesics) be excluded from influencing the effects and outcome of chest physiotherapy? Standardization of these therapies is unthinkable, yet pain relief is an important factor in the pre­ vention of respiratory complications after surgery and patient acceptance of chest physiotherapy and mobilization. The du­ ration, type, and frequency of interven­ tions in four different types of patients over an 8-hr period are described in Ap­ pendix IV. In some institutions there are interventions that do not occur at others. For example, if there is variability be­ tween one surgical ICU and another i n t h e morbidity an d mortality associated with the same surgical procedure, how can this discrepancy in standard of care be accounted for when comparing chest physiotherapy at the two institutions? These are just a few of the difficulties of evaluating chest physiotherapy in the ICU. The solution, for the most part, re­ mains unknown. Long-Term Follow-up

A 2-hr follow-up of respiratory changes after chest physiotherapy may be too long, since, allowing 1 hr for chest phys­ iotherapy and 2 hr for sampling and mea­ surement, variables are assumed to be

234

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

constant during a study period of about 3.5 hr. The number of interventions for four patients is shown in Appendix IV. The four patients studied demonstrate the care given to patients in a typical trauma ICU. They were observed from 8 A.M. to 4 P.M., and all interventions that caused alterations in cardiac or respira­ tory function or that made physiological study difficult were recorded. The mean number of clinically re­ quired interventions was 59 (Table 7.9). These interventions took, on average, 298 min or 5 min for each intervention. The nonintervention time was, therefore, limited to a mean of 93 min in the whole 8 hr. The longest intervention-free pe­ riod averaged 34.5 min, which repre­ sented only 9.1% of the total 8 hr. The duration of restricted access was 145 min on average. Restricted access was the term used to refer to situations that alter the ability to monitor (for example, pa­ tient 3 left the CCRU for hyperbaric 0, therapy and whirlpool debridement), change the hemodynamic status (for ex­ ample, patient 1 had dialysis that can cause considerable circulating volume shifts), or make comparative physiologi­ cal measurement impossible (for exam­ ple, patient 2 was, at times, rotated on the Roto-Rest bed). This alters hemodynam­ ics, ventilation/perfusion relationships within the lung, and respiratory mechan­ ics and function. The changes in heart rate, mean arterial blood pressure, mean P A pressure, ICP and cerebral perfusion pressure (CPP) in patient 4 are shown graphically in Appendix IV. The clini­ cally necessary adjustments to ventila­ tion and cardiac function exclude

acutely ill patients from lengthy study of the effects of chest physiotherapy on these parameters. An alternative to long follow-up of highly changeable parameters, such as pulse and cardiac output, is to record more nonspecific indicators. If these vari­ ables are recorded for a long period and in a large enough patient population, use­ ful data may be generated concerning the effectiveness of chest physiotherapy. In­ formation on daily 8 A.M. ABC, lung/tho­ rax compliance, and chest x-ray appear­ ance was, therefore, collected in 58 mechanically ventilated patients be­ tween August and October 1977. All re­ ceived chest physiotherapy and had suf­ fered multiple system injuries (see Appendix I, p. 352, Table A1.7 for defi­ nition of systems) and were admitted to the CCRU. The patients were divided into groups based on whether they had chest injury, head injury and other injury, pelvic frac­ ture, cervical spinal column injury, or extremity fracture (Table 7.10), Data were collected only in mechanically ven­ tilated patients and lhe numbers in each category from which the data were ob­ tained are shown at the top of Figures 7.18-7.20. Not all the data could be col­ lected on every patient each day. Chest X-Ray

The daily chest x-ray was assessed by using the following system: clear, 0; infil­ trate or plate-like atelectasis, 1; atelecta­ sis (segmental or lobar) or lung contu­ sion, 2; and pneumonia, 3. The results are shown in Figure 7.18. On admission, the

Table 7.9 Duration and Frequency of Interventions Causing Changes in Cardiac or Respiratory Function in Three Critically III Patients' Direct Intervention (min)

Nonintervention (min)

Patient 1 Patient 2

448 399

32 81

Patient 3

161

106

Patient 4

183

154

longest Period of Nonintervention (min)

Number of Interventions

359 (dialysis) 197 (rotation of Roto·

20 20

43 75

213 (hyperbaric chamber,

28

29

70

89

Restricted or No Access Time (min)

Rest bed)

whirlpool debridement) 35 (visitors)

"Full details of aU interventions appear in Appendix IV; each patient was observed a total of 480 min.

PHYSIOLOGICAL CHANGES FOLLOWING CHEST PHYSIOTHERAPY

of mechanical ventilation, between the chest-injured patients who received chest physiotherapy and the head-in­ jured patients who received routine turning and tracheal suctioning, but no chest physiotherapy (unless the chest x­ ray scored 2 or more). This suggests that chest physiotherapy prevented further deterioration in chest x-ray changes and that routine turning and suctioning did not.

Table 7.10 Injuries Sustained by 58 Long-Term Follow­ Up Patients Injury

Number of Patients

Average Number Body Systems Injured

26

3.0

31 16 29 13

2.9 2.7 2.7 2.1

Rib fractures; lung contusion Pelvic fracture Unconscious Extremity fracture Cervical spine fracture

PaOzlFIO,

The changes in PaO,/FIO, are shown in Figure 7.19. The patients with head in­ jury presented with the highest ratio (lowest intrapulmonary shunt), and the patients with chest and cervical spine in­ jury, with the lowest. After 1 week, two distinct groups emerged; the patients with chest, pelvis and extremity injury had low PaO,/FIO" and those with head and cervical spine injury had high PaO,/ FlO,. The differences between these two groups receiving chest physiotherapy are likely to be the result of the injuries sus­ tained by the different patient population rather than the chest physiotherapy. The

patients with chest injury presented with the highest score. and those with head injury, with the lowest. Two of the 19 pa­ tients with chest injury developed a pneumonia-like process that lasted 9 and 3 days, respectively. On average, in pa­ tients with multisystem injury, chest physiotherapy was successful in preven­ tion of major atelectasis or pneumonia (Fig. 7.18). Chest injury or pelvic fracture (or a combination) produced the highest score and the longest duration of chest x­ ray changes despite chest physiotherapy. There is a lack of difference, after 4 days

NUMBER OF PATIENTS ANALYSED

CHEST PELVIS HEAO EXTflEM C-SPINE

I.

,. ,. " I.

20

" 17 7

"

I>

12

I> 7

•

" " "

I> 7

" I>

"

16 •

" "

12 •

" •

• 7

10

10

12 •

10 •

10 >

7 >

• •

• • >

10

"

235

, ,

,

7

,

• 7

•

,

>

4

• •

•

• • 2

2 • 2

> >

>

> >

>

>

> 4

> •

> •

> >

2 >

>

>

4

•

>

2

__

•

___._

•

_____

•

_._._ • __

•

2 >

I

I

CHEST!CI PELVIS (PI HEAO ( H ) EXTREMITY (E1 C-SPINEICS)

2.0

1 .5

AVERAGE DAILY CHE§Zo�fY

I .0

PAft��T

0.5

o

c cs

"> -.:,-;. " .... .. f. l�,

"' -

! :>

Inspiratory Reserve Volume ( I RV I - - -

J���

Expiratory Reserve ( ERV I Volume Residual Volume ( RV I

>

;2 o uu � :;..=. .

;1 ..,

IRV

..

0

cU

-+ u i � ""

;; 'i � g :I u U

����

�:�-

....

;1 .. .. 0

B

..

...

- 191% .140'10

Figur. 8.7. Comparison of lung volumes in the normal (A) and the quadriplegic (8) patient. Note decreased VC. total lung capacity. and ERV in the quadriplegic patient. (... . Estenne. 1987: O. Fugl­ Meyer. 1 971 a:·. Haas. t 965).

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS SpLnal segment

Inspiratory

C

Expiratory

I

:rr :m: Jl[

265

Figure 8.8. (left). The respiratory muscles. Inspiratory muscles to the left and expiratory muscles to the right of the spinal segmental indi­ cator. 'Primary respiratory muscles (different opinions in the literature on whether these are primary or auxiliary). (From A. R. Fugl-Meyer: Scandinavia Journal of Rehabilita­ tion Medicine 3:141-150. 1 971 a.)

:l[ :l[ :llI[ :lZII[ Th

I :rr

�i

:m: c

I� !. , ,

I

•

I •

Jl[ I " ,

:l[

I

:lZ[ :llI[

!

:lZIJ[ IX X XI:

, " ,

i I I

•

o

T , •

i ! 1

�

t :

R !

!

I � �

;

III L

I ][ ][ Jl[ :l[

5

I :n: :m: Jl[ :l[

1970). Bake et a!. (1972) found that the abdominal contribution to quiet respira­ tion was 31% in normal subjects and 50% in the quadriplegic patient. Therefore, tetraplegic subjects have a smaller rib cage contribution and a greater diaphrag­ matic contribution to tidal volume than healthy subjects (Fugl-Meyer 1971b; Mortola and Sant'Ambrogio, 1978; Es­ tenne and DeTroyer, 1985, 1987). Acces­ sory muscle activity is variable in quad­ riplegics; this may affect the thoracic

component of ventilation that ranges from 22 to 90% of total ventilation (Mc­ Kinley et a ! . , 1 969). Estenne and De­ Troyer (1985) studied 20. C4-C7 quadri­ plegic patients between 10 days and 312 months after injury with EMG recordings of the parasternal intercostals and sca­ lene muscles. The degree of rib cage mo­ tion could not be predicted; there was no relationship between thoracic motion and duration of quadriplegia. Spastic or silent scalene EMG activity was associ-

266

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

ated with paradoxical upper rib cage AP motion. These authors conclude that quadriplegics have a very complex pat­ tern of muscle activity during inspiration and that coordination of the scalenes and diaphragm may be important. Three forces act upon the rib cage when the di­ aphragm contracts: a fall i n pleural pres­ sure. a rise in abdominal pressure. and a force on the insertion of the diaphragm elevating the lower ribs. Pulmonary Function

Vital capacity has been shown to dou­ ble within 3 months of injury and con­ t i n ue to i ncrease spontaneously 4-10 months after quadriplegia (Ledsome and Sharp. 1981; Haas et al.. 1986; Axen et al.. 1985). Proportional i ncreases in inspira­ tory capacity and total lung capacity (TLC) and maximum inspiratory pres­ sure (Plm,,) also occur in the acute stage (McMichan et al.. 1980; Haas et al.. 1986). During the chronic stage. vital capacity i ncreases and functional residual capac­ ity decreases while TLC remains the same. Absolute improvement i n vital ca­ pacity cannot be predicted by pulmonary function tests. neurologic examinations. or muscle function evaluations per­ formed in the early stage of recovery (Axen et al.). Postural Dependence. Pulmonary function varies with body position and the use of abdominal binders in the spon­ taneously breathing quadriplegic patient. Unlike normal subjects whose vital ca­ pacity decreases by 7.5% in the supine position (Allen et al. . 1 985). patients with cervical cord transection have a de­ creased vital capacity. tidal volume. and inspiratory capacity. increased residual volume (RV). and decreased ventilation in the lung bases when changing from the supine to seated position (Maloney. 1979; Haas et al.. 1965: Fugl-Meyer. 1971a; Bake et al.. 1972; McMichan et al.. 1980; Estenne and DeTroyer 1987). Vital capacity is increased by the 20' head­ down position (Cameron et al.. 1955). Total l u ng capacity is smaller i n the su­ pine position (Estenne and DeTroyer. 1987) probably because of a reduction i n R V . although inspiratory capacity (IC). VC. and tidal volume increase (Maloney.

1979). Reduced residual volume is thought to be related to the effect of grav­ ity on the abdominal contents and not an abnormal i ncrease in intrathoracic blood volume (Estenne and DeTroyer. 1987). McMichan and colleagues associated the i ncreased lung volumes with shortened diaphragmatic descent and lack of ab­ dominal rebound. Increased paradoxical inward movement of the lateral chest wall is also noted in the supine position when compared to silting (Moulton and Silver. 1 970; Mortola and Sanl' Ambro­ gio. 1978; Estenne and DeTroyer). The re­ duced lower rib cage expansion in the supine position is thought to be due to the i ncrease in abdominal compliance that occurs with assuming this position (Estenne and DeTroyer). Cardiac function is also affected by changes in position after cervical cord transection or spine i n jury. Rapid changes in body position for the acute quadriplegic patient during spinal shock may cause marked changes in cardiac function. Head elevation of greater than 20' may cause a sudden decrease in car­ diac filling pressures. a resulting fall in cardiac output. and even cardiac arrest. Similarly. sudden head-down positioning may cause a rise in cardiac filling pres­ sures. Because of loss of sympathetic car­ diac i n nervation associated with lesions above the T1 level. the steep head-do W'n position may precipitate acute myocar­ dial failure with pulmonary edema. Therefore. in the early stages of acute quadriplegia. these movements should be performed with careful monitoring of arterial and venous pressures. Ace wraps around the lower extremities. a G suit. or M ASTrousers may be used to minimize orthostatic hypotension until vasomotor control is established. Abdominal Binders. Abdominal bind­ ers are used to align the abdominal con­ tents under the diaphragm. thus improv­ ing respiratory function both in the acute and chronic phases of quadriplegia. When comparing eight C5-C7 quadriple­ gic subjects 1-456 months after injury to five normal subjects. McCool et al. (1986) demonstrated that abdominal binding in­ creased IC. TLC. and decreased FRC in the quadriplegic subjects. FRC and TLC decreased in normal subjects in all three

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS

_

positions tested [supine, seated, and tilted head-down 37·). The greatest im­ provement i n inspiratory capacity for quadriplegic patients was in the seated and tilted positions, because of the nor­ mal lengthening of the diaphragm that occurs in the supine position. Maloney studied 15 quadriplegic patients 1 year postinjury. lt was found that wearing a corset in the sitting [not supine) position improved lC, VC, and tidal volume. The authors conclude that the increased ab­ dominal pressure associated with binder use improves rib cage expansion. Gold­ man and colleagues [1986) also demon­ strated an increase in transdiaphragmatic pressure and VC with abdominal binding i n the sitting position, although VC was not altered i n the supine position. Es­ tel).ne and OeTroyer [1987) noted that ab­ dominal binding abolished the postural dependence of RV in the supine position, the effect on TLC is not mentioned. Imle et a1. [1986) studied the affects of abdom­ 'inal binding on acute quadriplegics and also documenled Ihat VC was unchanged with binding in the supine position. Binder type and placement may be cru­ cial in demonstrating improvement in pulmonary function. Elastic binders wrapped tightly around the abdomen, extending over the iliac cresls to Ihe pubis, are preferred [McCool et aI., 1986; Goldman et aI., 1986). The binder should not be positioned more cranially than the floating ribs because of interference with epigastric rise during inspiration [Alva­ rez et aI., 1981). It is the author's experi­ ence that binders placed below the ante­ rior superior iliac spine are more prone to tissue breakdown, particularly in the sit­ ting posi tion. When improperly donned, thoracic mobility may be impaired. Bind­ ers with an orthoplast front are not as effective as conventional binders in im­ proving end inspiratory tidal and trans­ diaphragmatic pressure during maximal sniffs [Goldman, et a1.). The overall benefits of wearing an abdominal binder remain unknown be­ cause of the decrease i n FRC that may impair gas exchange. Improvement in VC enhances cough ability, yet decreased FRC leads to alveolar collapse. The au­ thors have found binders clinically ben­ eficial for some quadriplegic palienls

267

both to decrease respiratory rale and in­ crease VC and tidal volume when used i n supine a n d sitting positions. Pulmonary function testing may be performed when using binders, particularly with supine positioning. Binders are worn until pul­ monary function ceases to improve with their use, or breathing fails to appear eas­ ier during functional activities. Chest Physiotherapy Treatment

Reduced VC, restricted deep breath­ ing, and cough, together with an inability to change body position, lead to secretion retention. Quadriplegic patients are, therefore, very susceptible to pulmonary complications [McMichan et a1.). In order to improve respiratory management, chest physiotherapy treatment, including breathing exercises and specific active and passive range of motion exercises are necessary [see pp. 163). Chest physio­ therapy is reported to be highly success­ ful in reducing pulmonary complications in these patients [McMichan et a1.). Treatment is instituted prophylactically and continued, with emphasis on any areas showing radiological involvement. No benefit from the use of bronchodila­ tors for Ihe quadriplegic patient was doc­ umented by Fugl-Meyer [1 976). It is our opinion thai adventitial breath sounds, including wheezing, are often a result of retained secretions, since they clear with chest physiotherapy. Both the mechani­ cally ventilated and spontaneously breathing quadriplegic patient require immediate attention to prevent atelecta­ sis and pneumonia. Quadriplegic patients should have chest physiotherapy performed during weaning from mechanical ventilation. The authors are i n agreement with Wicks and Menter [1 986) that IMV is not partic­ ularly beneficial for Ihese patients. Re­ tained secretions are removed prior to a weaning session to enhance gas exchange and decrease the work of spontaneous breathing. This is particularly important for the quadriplegic who may lack the necessary intercostal and accessory mus­ cle strength to decrease diaphragmatic muscle fatigue [Lerman and Weiss, 1987). While spontaneously breathing cough as­ sistance and breathing exercises are con-

268

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Figure 8.9. The physical therapist clinically assesses a quadriplegic patient prior to performing per­ cussion over the posterior seg­ ments of the lower lobes. The turning frame allows head-down positioning .

tinued for secretion removal, ventilatory muscle strengthening, and relaxation. Postural Drainage, Percussion, and Vibration

Both the turning frame and a standard bed permit chest physiotherapy in 7 of the 11 bronchial drainage positions. Per­ cussion and vibration may be performed over appropriate lung segments (Fig. 8.9). Thoracic excursion i n varied positions that include more than one plane of mo­ tion may prevent cavus deformity and flaring of the lower rib cage, particularly in children (Massery, 1987). Position changes for quadriplegic patients should be made carefully by experienced health care personnel because of their effect on cardiac and pulmonary function. I t is our opinion that a standard bed and turning frame are superior to the kinetic bed for care of the quadriplegic patient. ComFigure 8.10. Cough assistance for a quadriplegic patient is achieved by placing a towel over the abdomen and the therapist ap­ plying even pressure during expiration.

pared to the kinetic bed the turning frame and standard bed allow more op­ timum postural drainage of the lower and middle lobes and better positioning for exercises to enhance early rehabilitation and prevent contractu res. The manufac­ turers claims of improved pulmonary function and decreased tissue break­ down and contracture formation with the kinetic bed are not substantiated by our clinical practice. Skin breakdown occurs on the heels and buttocks and shoulder contractures may develop when quadri­ plegic patients are managed on a kinetic bed. See Chapter 3 for details regarding specialty beds and positioning the spinal injury patient. Quadriplegic patients may be taught to cough while lying supine"on their side, and prone (Fig. 8.10). The turn­ ing frame may be positioned in the head­ up, head-down, prone, and supine posi­ tions (Fig. 8.11). Adequate diaphragmatic excursion can be obtained for patients

269

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS

Figure 8.11. One person may raise or lower th� head of the Stryker frame. positioned prone on a turning frame by placing a roll under the iliac crests (Fig. 8.12). Well-padded straps may be used to prevent the patient from slipping and los­ ing traction while in the head-down po­ sition (Fig. 8.13). If straps are used they should be released immediately follow­ ing treatment to prevent tissue break­ down or circulatory occlusion. Cameron et al. (1955) describe padded boots to achieve the same effect. Once adequate traction is assured. both hydraulic and manual frames may be positioned for postural drainage. During rehabilitation. the prone on elbows position may be used on a bed or mat to strengthen neck accessory muscles while limiting and re­ sisting diaphragmatic excursion. The philosophy of the neurosurgical or orthopedic staff will determine when

surgical intervention and use of a vest for spinal stabilization are indicated. A halo vest is preferred to the use of a body cast. Manual chest physiotherapy techniques are more easily performed because the straps can be u n fastened and vest opened (Fig. 8.14). The jacket is u n fastened on both sides and the back raised to allow percussion over the posterior basal seg­ ments of the lower lobes. Padding can be added for patient comfort. two or three staff members are required to position the patient prone i n a vest. Jacket straps should be closed prior to changing the patients position. Chest physiotherapy is also performed when a Yale brace is re­ q u i red for spinal stabilization (Fig. 8.1 5). When body casts are used windows should be cut to expose the chest wall (Fig. 8.16). Casts compromise respiratory function. i n terfere with chest-ray inter­ pretation. and cause pressure sores. I n patients w i t h quadriplegia and copious secretions. prolonged periods of the head-down position may be necessary to assist secretion drainage into the oro­ pharynx. I n a minority of patients. 90120 min of drainage is required for one or two treatments before clinical clearance is noted. This is especially the case i n spontaneously breathing patients wh o d o n o t cough well. Cough

Due to abdominal and i n tercostal mus­ cle weakness. cough is often severely im­ paired in the quadriplegic patient. If the patient is taught to take as large an inspi­ ration as possible. followed by a forceful expiration during which the abdomen is

Figure 8.12. A roll under the iliac crests allows anterior diaphragmatic excursion during prone positioning on a turning frame .

270

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Figure 8.13. Kerlex and abdomi­ nal dressing pads are easily ob­ tained in the ICU and can be util­ ized to prevent the patient on a turning frame from slipping when placed head·down.

Figure 8.14. A quadriplegic pa­ tient is receiving chest percussion over the left lower lobe after the vest used for spinal stabilization is opened and taped to the siderails of the bed. The vest was opened after proper patient positioning.

Figure 8.15. When a Yale brace is used for spinal stabilization. the straps may be unfas­ tened to allow chest physiotherapy treatment.

Figure 8.16. When body casts are used for spinal stabilization in the quadriplegic patient. windows must be cut to expose the chest wall.

271

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS

Table 8.1 Comparison 01 Air Flow, Duration, and Volume 01 Cough in Quadriplegic and Normal Subjects

Peak Air Flow (liters/sec)

Duration of Cough (sec)

Volume of Cough (liters)

Resistance to Air Flow (Peak Flow) (cm H,O/ liters sec)

7.09 4.54

1 .09 2.3

3.14 2.91

12.50 2.59

Normal Quadriplegic

supported. a more effective cough may be achieved (Fig. 6.10). The patient can be taught to perform this maneuver inde­ pendently. Huffing is also helpful (see p. 1 62). Siebens et al. (1964) found abnormal volume and pressure changes during coughing in three male quadriplegic pa­ tients with C5 and C6 spinal cord tran­ secfiDns (Table 6.1). Flow and resistance were decreased compared to those values in three healthy men. Breathing Exercises

Breathing exercises to increase tidal volume and assist coughing are advo­ cated for the nonintubated quad riplegic patient. Quadriplegic patients with a vital capacity less than 1 .000 ml usually require mechanical ventilation (Wicks and Menter, 1986). Because alternating periods of rest and exercise improve pul­ monary function in some patients (Braun et aI., 1963), the authors use breathing exercises during periods of spontaneous breathing while weaning from mechani­ cal ventilation. Exercising the intact res­ piratory muscles such as the diaphragm, sternocleidomastoid, levator scapulae, platysma. and trapezius may increase thoracic and abdominal excursion. there­ fore increasing tidal volume (Cullmann. 1976; McMichan et al.. DeTroyer and Heilporn. 1 980; Wetzel et al.. 1 985). Breathing exercises most often taught to quadriplegic patients include active and resistive diaphragmatic breathing. summed breathing exercises. inspiratory muscle training. and glossopharyngeal breathing. Fifteen chronic quadriplegics improved VC after 7-12 weeks of incen­ tive spirometry and arm ergometry (Walker and Cooney. 1967). Active dia­ phragmatic breathing exercises and in ­ spiratory muscle training are described in Chapter 3. Breathing exercises are initially taught

to the quadriplegic patient in the supine position because of the associated in­ crease in vital capacity. Resistive dia­ phragmatic exercises are achieved by placing dish or cuff weights over the epi­ gastric region (Fig. 8.17). Inspiratory ca­ pacity and vital capacity are measured to determine the maximum weight used during training while achieving a full epigastric rise. During breathing exer­ cises and pulmonary function testing. nose clips are worn u n less the patient is tracheally intubated. In the acute phase. for the spontaneously breathing quadri­ plegic patient the authors advocate dia­ phragmatic breathing exercises for 40 re­ petitions twice daily five days per week or inspiratory muscle training for 15 min twice daily. Ciesla et al. (1 989) compared the effec­ tiveness of abdominal weight training to inspiratory muscle training on 29 acute C4-C7 quadriplegic subjects. FVC. IC, MVV. PEFR. and Plm" were measured be­ fore and after 3 weeks of training. No sig­ nificant difference was found between

Figure 8.17. A quadriplegic patient performs diaphragmatic progressive resistive exercises with dish weights. Weights are added until the patient's inspiratory capacity is greater than or equal to baseline measurements.

272

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

these treatment modalities, although sig­ nificant improvement (p < .05) was noted in both subject groups. Glossopharyngeal Breathing

Glossopharyngeal breathing (GPB) is recommended for the patient with quad­ riplegia. The technique can be broken into four steps [Fig. 8.18 A-C). Oail and Affeldt (1955) studied GPB in patients with poliomyelitis. Of 100 patients who were taught GPB, 69 used i t to assist nor­ mal breathing, and 31 used it to assist speech and stretch the chest to help coughing. Forty-two of these patients found that GPB freed them of the need for respiratory support. Ardran et al. (1959) found that the GPB rate varied from 60 to 200 times/min in individuals with poliomyelitis. Those pa­ tients who could swallow normally were capable of GPB; those with palatal and la­ ryngeal weakness were sometimes capa­ ble of GPB if a nose clip was used. It may not, therefore, be necessary to close the larynx for GPB. Affeldt et al. found no re­ lationship between the polio patient's proficiency a t GPB and the severity of respiratory muscle paralysis. VC was in­ creased from 0.28 to 2.40 liters, with GPB ranging from 14 to 48 strokes/min, and normal arterial blood gases were main­ tained. The mean increase of VC ranged from 11 to 50% of the predicted normal, in seven patients. Metcalf (1 966) and Montero et al. (1967) specifically studied quadriplegic patients. Metcalf found that vital capacity was increased from 60 to 81% of normal by means of GPB. Montero and colleagues believed that if 700-1,000 ml of air could be added to a patient's VC with 10-20 glossopharyngeal gulps, the technique was then mastered. This vol­ ume provides sufficient supplemental air for effective coughing and secretion clearance. Many quadriplegic patients who require respiratory support undergo tracheostomy to reduce the problems as­ sociated with long-term translaryngeal intubation. Therefore, GPB is limited to extubated patients or those without an inflated tracheostomy tube cuff [G. T. Spencer, personal communication). "Summed breathing" may also in­ crease tidal volume. This is carried out

by encouraging the patient to take sev­ eral quick, shallow but cumulative breaths before expiration. The patient can gradually increase the volume of in­ spired air once this is mastered. Fugl-Meyer [1 971b) devised a manu­ ally operated pump and valve system that was used as a passive breathing ex­ ercise and increased total lung capacity 14% in quadriplegic patients. Productive coughing was improved. This was most effective with patients in the sit t ing position. In summary, prophylactic chest phys­ iotherapy is important for the patient with acute quadriplegia. Emphasis on in­ creasing VC and cough efficacy are essen­ tial in the spontaneously breathing pa­ tient. Exercising the remaining accessory muscles, performing diaphragmatic, in­ spiratory resistive, glossopharyngeal, and summed breathing, or the use of a pump and valve system may all improve pul­ monary function. I n order to maintain rib cage mobility, these exercises are started as soon as possible following injury. Ac­ tivities of daily living may eventually re­ place the need for chest physiotherapy and breathing exercises. PATIENTS WITH CHRONIC LUNG DISEASE

Evidence of the benefits of chest phys­ iotherapy for patients with chronic ob­ structive lung disease is limited. Anthon­ isen et al. (1964), Petersen et al. (1967), March (1971), Newton and Bevans (W78), Newton and Stephenson (1978), an il Ol­ denburg et al. (1979) were unable to dem­ onstrate improvement in pulmonary function or sputum clearance with chest physiotherapy. May and Munt (1979) found postural drainage and percussion effective in augmenting the volume of expectorated sputum, but this did not produce significant alterations in air flow or gas exchange. Campbell et al. (1975) found a fall in FEV, when percussion was added to postural drainage and coughing. The patient who is subacutely ill and has chronic lung disease with retained secretions is encouraged to become in­ dependent in activities of daily living, as opposed to having a vigorous chest phys­ iotherapy regime implemented. Pulmo-

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS

273

CJ

�,gC.LAORSYENXD o

..

A

0 (I

3

Figure 8.18. (A) Steps of GPB: (1) The mouth and throat are filled with air; the tongue, jaw and larynx are depressed. (2) The lips are closed, and the soft palate is raised to trap air. (3) The larynx is opened; the jaw and floor of the mouth and larynx are then raised. With repeated motion of the tongue, air is forced through the opened larynx into the trachea. (4) The larynx is closed, and air is trapped in the trachea and lungs. (From C. W. Dail and J. E. Affeldt: Journal of the American Med­ ical Association 1 58:445-449, 1 955.) (8) GPB steps 1 and 2: The jaw and larynx are depressed; the soft palate is raised. (C) GPB steps 3 and 4: Air is forced through the opened larynx into the trachea. nary rehabilitation that includes patient education and exercise testing and train­ ing are indicated. Exercise training may include PaO, measurements during ex­ ercise and the use of supplemental oxy-

gen; inspiratory muscle training (IMT) may also be indicated (pp. 1 2 2 ) (Butts, 1981: Stein et aI., 1982; Ries et aI., 1983). Whether (IMT) alone improves respira­ tory muscle endurance in patients with

274

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

chronic lung disease is unknown (Bel­ man and Sieck, 1 982). lMT has not been shown to increase exercise tolerance when compared to a pulmonary rehabil­ itation program (Casaburi and Wasser­ man, 1 986), although Ries et a! . ( 1 986) demonstrated improvement in ventila­ tory muscle endurance and exercise per­ formance when IMT was compared to a walking program. See Chapter 3 for de­ tails regarding IMT. The major indica­ tions for chest physiotherapy in the pa­ tient with chronic lung disease are excess sputum production, exacerbations of the disease resulting in immobility, or major abdominal surgery or trauma. Cochrane et a!. (1977) showed reduced air flow obstruction, and Bateman et a!. (1 979) and Sutton et a!. ( 1 982) showed increased clearance of radioactive poly­ styrene particles from central and pe­ ripheral airways following chest physio­ therapy in patients producing regular daily sputum. Similarly, Feldman et a!. demonstrated increased expiratory air flow i n ten patients with chronic bron­ chitis up to 45 min after postural drain­ age, percussion, and vibration in six positions. After surgery, spontaneously breathing and mechanically ventilated patients with chronic sputum-producing lung dis­ ease are likely to retain secretions as a re­ sult of immobility, pain and the use of dry anesthetic gases. A l tered pulmonary function before surgery puts these pa­ tients at a greater risk for respiratory complications. Prophylactic chest phys­ iotherapy, which includes the forced ex­ piration technique, is, therefore, indi­ cated. Chronic lung diseased patients may need to be coaxed into the necessary postural drainage positions. If retained secretions i nterfere with gas exchange, tolerance to treatment usually improves as treatment is continued. The sponta­ neously breathing patient who becomes more dyspneic with treatment may ben­ efit from relaxation and gentle condition­ ing exercises i n the sitting position. This may aid muscular relaxation and help conserve energy needed for effective deep breathing and coughing. As the pa­ tient's shortness of breath improves, pos­ tural drainage may then be administered. If low rates of intermittent mandatory ventilation (IMV) are in use. the patient

with chronic obstructive lung disease or cardiac disease may become dyspneic when placed in the head-down position. Increasing the mandatory ventilation rate or fractional inspired oxygen con­ centration, using controlled mechanical ventilation, or pressure support may allow these patients to tolerate better the head-down positions necessary for pos­ tural drainage of the middle and lower lobes.

ASTHMATIC PATIENTS

Physiotherapy for patients with asthma consists of breathing retraining exercises, physical conditioning, and postural exercises (Livingstone, 1952; Wood et a!., 1970; Mascia, 1976; Landau, 1 977). Breathing exercises are often used to reduce anxiety and relieve dyspnea (Freedberg et a!., 1987). Singh (1987) studied 12 asthmatics with nocturnal wheeze. In a controlled study statistically significant increases i n PEFR were noted following breathing exercises using a "pink city lung exerciser." This device maintained a 1 : 2 inspiratory-expiratory ratio, similar to that obtained with dia­ phragmatic breathing exercises (see Chapter 3). Postural drainage with per­ cussion and vibration is only necessary when the asthmatic patient has excess mucus production or secretion retention which is present after physical condition­ ing, breathing, or postural exercises. The spontaneously breathing asth­ matic with retained bronchial secretid)'ls may require breathing exercises or relax­ ation before postural drainage. The pa­ tient is positioned to promote relaxation of the upper chest and shoulder girdle musculature. Relaxation in several posi­ tions, such as sitti ng, lying on the side, and standing, should be incorporated when possible into the treatment. Pos­ tural drainage with percussion and vibra­ tion often cannot be tolerated unless the patient is relaxed. Chest physiotherapy is directed at the speCific areas of segmental atelectasis (Wood et a!.; McKaba, 1 976). Huber et a!. (1974) showed up to a 40% increase in forced expiratory volume, 30 min following percussion and vibration in 1 1 asthmatic children with mild-to­ moderate airway obstruction. This sug-

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS gesls that chest physiotherapy and secre­ tion removal decrease bronchospasm. Pa­ tients with asthma, hospitalized for treatment other than asthma, usually tol­ erate chest physiotherapy. The head­ down position may be used when it is in­ dicated. This is exemplified by the fol­ lowing case study. Case History 8.1. A 1 5-year-old female was admitted to the trauma center after an auto ac­ cident in which she was a backseat passenger. The patient sustained a fractured right pubis and sacroiliac jOint, a ruptured bladder, liver lacerations, a serosal tear of the rectum, and a retroperitoneal hematoma. Her past medical hl story was noncontributory except for a his-

Figure 8.19.

275

tory of asthma and shortness of breath on exertion. Following admission, the patient underwent laparotomy, and a suprapubic cystostomy was performed. After surgery the patient showed radiological evidence of a right upper lobe at­ electasis that cleared with chest physiother­ apy. The patient was given prophylactic chest physiotherapy every 4 hr because of her his­ tory of asthma. This was supplemented in the evening and at night by the nurSing staff. There were no turning restrictions, and the patient's chest x-ray remained clear until the fifth day after surgery when, due to concern over the pelviC fracture (Fig. 8.19A), turning was limited to lying on the left side only. Two days later the patient developed atelectasis of the left lower lobe (Fig. 8.19B) and an associated tempera-

(A) Pelvic fracutres include a fractured right pubis and sacroiliac joint. (B) Chest x­ ray showing left lower lobe atelectasis. (C) The left lower lobe atelectasis has cleared following 45 min of chest physiotherapy.

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

276

ture spike to 1 02°F. Turning was again permit­ ted, and chest physiotherapy was given. Treat­ ment consisted of postural drainage and vigorous percussion. vibration and assisted coughing with the patient in the head-down po­ sition. Side-lying prone and supine positions were included while the patient was in the head-down position. Treatment lasted 45 min, at which time the physiotherapist believed that the lungs were clear on auscultation, except for some wheezing which was apparent since admission. Repeat chest x-ray revealed complete clear­ ing of the left lower lobe atelectasis (Fig. B.19C). Temperature decreased to 99°F. A turning frame was subsequently used to man­ age both the patient's pelvic fracture and her pulmonary condition. The patient in status asthmaticus does not usually require chest physiotherapy initially (British Medical Journal edito­ rial, 1972). However, following medical treatment, breathing control and chest physiotherapy may be instituted to assist secretion removal (Wood et aI., Webber, 1973). The i ntubated asthmatic patient is especially prone to secretion retention as a result of bronchospasm, immobility, decreased ciliary activity, and i nterfer­ ence with the normal cough mechanism. Therefore, routine turning and suction­ ing are performed. In addition, chest physiotherapy treatment of any areas of the l ungs with clinical or radiological ev­ idence of secretion retention may be helpful in reducing bronchospasm. Re­ tained secretions may cause airway ob­ struction resulting in wheezing and should, therefore, be removed. Treat­ ment of the asthmatic should be guided by sputum production and patient toler­ ance. Chest physiotherapy treatment of the asthmatic patient should follow pre­ scribed bronchodilator administration whenever possible.

SUMMARY

Chest physiotherapy when adminis­ tered to different patient populations has variable effects. In neonates, chest phys­ iotherapy appears to be more hazardous than in adults, and there is evidence of hypoxemia associated with suctioning, handling. and chest physiotherapy. Con­ versely, removal of secretions, mobilized

by manual techniques, may be more im­ portant in children because of their small airways which are more easily occluded with retained secretions and the in­ creased number of mucus glands. Pa­ tients with cystic fibrosis respond favor­ ably to chest physiotherapy, which includes the forced expiration technique, although further research is needed to determine if the effects of therapy vary with the severity of the disease and gen­ eral physical conditioning. Patients with spinal cord injury and neurological defi­ cit involving intercostal and abdominal muscle activity require prophylactic chest physiotherapy to assist mobiliza­ tion and expectoration of secretions. In the quadriplegic patient, specific breath­ ing exercises are beneficial to improve coughing and secretion clearance. Ab­ dominal binders may improve vital ca­ pacity and cough ability. Unconscious patients with closed head injuries are prone to i ncreased retention and stagna­ tion of secretions because of immobility and poor cough. These patients usually tolerate chest physiotherapy in the head­ down position despite transient increases i n ICP. The primary indications for chest physiotherapy in patients with obstruc­ tive lung disease or asthma occur during acute exacerbations and after surgery or trauma. In chronic stages of these dis­ eases, the benefit of chest physiotherapy is not established.

References Affeldt IE. Dail CW, Collier CR, Fa" AF': Glclsso­ pharyngeal breathing: Ventilation studies. J AppJ Physio1 8:1 1 1 -1 1 3, 1 955 Allen SM. Hunt B, Green M. Fall in vital capacity with posture. Bf J Dis Chest 79:267-272, 1985 Alvarez SE. Peterson M. Lu nsford BR: Respiratory treatment of the adult patient with spinal cord in­ jury. Phys Ther 61:1 737-1745. 1981 Angus E. Thurlbeck W: Number of alveoli in the human lung. J Appl Physio/ 32(4):483-485, 1912 Anthonisen p, Riis P. Sogaard·Anderson T: The value of lung physiotherapy in tho treatment of acute exacerbation in chronic bronchitis. Aclo Med Scand 175:715-719. 1 964 Ardran GM. Kelleher WHo Kemp FH: Ci neradio­ graphic studies of glossopharyngeal breathi ng. Br J Radio 32:322-328. 1959 Auer LM. Cell C . Richling B. el al.: Predicting lethal outcome after severe head injury-a computer assisted analysis of neurological symptoms and laboratory values. Aclo Neurochir 52:225-238. 1 980

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS Axen K. Pineda H. Shunfenthal H. Haas F: Dia­ phragmatic function [ollowing cervical cord in­ jury: Neurally mediated improvement. Arch Phys Med Rehobil 66:219-222. 1985 Baigelman W, O'Brien I: Pulmonary effects in head trauma. Neurosu rgery 9(6):729-740, 1981 Bain J. Bishop J . Olinsky A: Evaluation of directed coughing in cystic fibrosis. 8r. / Dis ChesI 82:138148. 1 988 Bake B. Fugl.Meyer AR. Gimby C: Breathing pal­ terns and regional ventilation distribution in lelrapiegic patients and in normal subjects. Clio Sci 42:117-128, 1972 Barnes CA, Asonye UO. Vidyasaqar 01: The effects of bronchopulmonary hygiene on Pt(;OZ values in critically ill neonates. Cril Care Med 9(12):819822. 1981 Bateman IRM. Newman SP, Daunt OM, Pavia D. Clarke SW: Regional lung clearance of excessive bronchial secretions during chest physiotherapy in patients with stable chronic airways of ob­ struction. La ncel 1 :294-297. 1979 Bedford RF, Persing IA. Pobereskin L. Butler A: Lo­ docaine or thiopental for rapid control of int racra­ nial hypertension? Anesth Analg (Cleve) 59:435437. 1980 Bellamy R. Pills FW. Stauffer ES: Respiratory com­ plications in traumatic quadriplegia. I Neurosurg 39:596-600. 1973 Belman MJ, Sieck G: The ventilatory muscles. fa­ tigue. endu rance and traini ng. Chest 82(6):761766. 1 982 Braun NM. Faulkner J. Hughes R, Roussos C. Sahgal V: When should the respiratory muscles be ex­ ercised? Chest 84(1 }:76-84. 1983 Brimioulle S. Moraine JJ, Kahn RI: Passive physical therapy and respiratory therapy effects on intra­ cranial pressure (Abstract). Crit Core Med 16(4 ):449. 1 988 Brifish Medical lournal Editorial: Treatment of sta­ tus asthmaticus. Br Med J 4:563-564. 1972 Burnard ED, Grattan-Smith P. et a1.: Pulmonary in­ sufficiency in prematurity. Ausl Paedialr J 1:1238. 1965 Butts J: Pulmonary rehabilitation through exercise and education. GVP 17-61. December-January. 1981 Cameron GC. Scott JW, lousse AT. Bolterell EH: Di­ aphragmatic respiration in the quadriplegic pa­ tient and the effect of position on his vital capac­ ity. Ann Surg 141 :451-456. 1955 Campbell AH. O'Connell JM. Wilson F: The effect of chest physiotherapy upon the F'EV. in chronic bronchitis. Med I Aust 1 :33-35. 1975 Campbell EJM. Agostoni E. David IN: The Respira­ tory Muscles: Mechanics and Neural Control. p 46. WB Saunders. Philadelphia. 1970. Casaburi R. Wasserman K: Exercise training in pul­ monary rehabilitation. N Engl I Med 314:1 5091 5 1 1 . 1986 Cheshire DIE: Respiratory management in acute traumatic tetraplegia. Paraplegia 1 : 252-261. 1 964 Ciesla N. Simpson N. Derrickson I . Salmon M: A comparison of two different breathing exercises for quadriplegic patients (abstract). Phys Ther 69:393. 1 989 Cochrane M. Webber BA. Clarke SW: Effects of spu­ tum on pulmonary funclion. Br Med I 2: 1 1811183. 1 977

277

Crane L: Physical Therapy in neonates with respi­ ratory dysfunction. Phys Ther 61(12):1764-1 773. 1981 Crane L: PhYSical Therapy for the neonate with res­ piratory disease. In Ca rdiopul monary Physical Therapy. edited by S Irwin and J Tecklin. pp. 305310. Mosby. SI. Louis. 1985 Cropp GI, Pullano TP. Cerny FJ. Nathanson IT: Ex­ ercise tolerance and cardiorespiratory adjust­ ments at Peak Work Capacity in Cystic Fibrosis. Am Rev Respir Dis 126:211-216. 1982 Currie DC. Munro C. Gaskell D. Cole I: Practice. problems and compliance with postural drainage: A survey of chronic sputum producers. Br I Dis Chesl 80:249-253. 1 986 Dail CWo Affeldt IE: Clinical aspects of glossopha· ryngeal breathing. lAMA 1 58:445-449. 1955 Danford O. Miske S. Headley I. Nelson RM: Effects of routine care procedures on transcutaneous ox· ygen in neonates: A quantitative approach. Arch Dis Childhood 58:20-23. 1983 Danon J . Druz WS. Goldberg NB. Sharp IT: Function of the isolated paced diaphragm and the cervical accessory muscles in CI quadriplegics. Am Hev Respir Dis 1 1 9:909-919. 1979 DeCesare I: Physical therapy for the child with res­ piratory dysfunction. In Cardiopulmonary Physi­ cal Therapy. edited by S Irwin and I Tecklin. pp. 334-338, Mosby. SI. Louis, 1985 DeCesare J. Babchyck BM. Colten HR. Treves S: Ra­ dionuclide assessment of the effects of chest physical therapy on ventilation in cystic fibrosis. Phys Ther 62(6):820-825. 1982 Denton R: Bronchial secretions in cystic fibrosis: The effects of treatment with mechanical percus­ sion vibration. Am Hev Respir Dis 86:41-46. 1 962 Desmond KI. Schwenk WF. Thomas E. Beaudry PH. Coates AL: Immediate and long-term effects of chest physiotherapy in patients with cystic fibro­ sis. J Pediolr 103:538-542. 1983 DeTroyer A. Heilporn A: Respiratory mechanics in quadriplegia: The respiratory function of the in­ tercostal muscles. Am Rev Respir Dis 122:591599. 1 980 DeTroyer A. Kelly S: Action of neck accessory mus­ cles on rib cage in dogs. J Appl Physio/ 56:326332. 1984 DeTroyer A, Estenne M. Ninane V: Rib cage me­ chanics in simulated diaphragmatic paralysis. Am Rev Respir Dis 1 3 2:793-799. 1985 DeTroyer A. Estenne M . Heilporn A : Mechanism of active expiration in tctraplegic subjects. N Engl I Med 314:740-744. 1 986 Dhande VG. Kattwinkei l , Darnall RA: Prone posi­ tion reduces apnea in preterm infants. Pediafr Res 1 6(2):285. 1982 Doershuk CF. Fisher BI. Matthews LW: Pulmonary physiology of the young child. In Pulmonary Physi% gy o! the Fetus Newborn and Child. edited by EM Scarpelli, pp. 167-169. Lea & Febiger. Phil­ adelphia. 1975 Estenne M. DeTroyer A: Relationship between res­ piratory muscle elect romyogram and rib cage mo­ tion in tetraplegia. Am Rev Respir Dis 1 32:53-59. 1985 Estenne M . DeTroyer A: Mechanism of the postural dependence of vital capacity in tetraplegic sub­ jects. Am Rev Respir Dis 135:367-371. 1987 Etches MB. Scott B: Chest physiotherapy in the

278

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

newborn: Effect on secretions removed. Pediol­

ries 62:713-715. 1978 Fan LL. Flynn JW, Pathak DR. Madden \VA: Predic­ tive value of stridor in detecting laryngeal injury in ext ubated neonates. Crit Care Med 10(7):453455. 1 982 Feldman J. Changes in maximal expiratory flow volume cu rves after postural drainage in patients wilh cystic fibrosis or chronic bronchitis (ab­ stract). Am Rev Respir Dis 1 1 3:272. 1976 Feldman J. Traver CA. Taussig LM: Maximal expi­ ratory flows after postural drai nage. Am Rev Res­ pir Dis 1 1 9, 239-245. 1979 Finer NN. Boyd J: Chest physiotherapy in the neo­ nate: A controlled study. Pedialrics 61:282-285. 1978 Finnie NR: Handling the Young Cerebral Palsied Child 01 Home, pp 24-28. William Heinemann Medical Books, London, 1971 Fisher 0, Frewen T, Swedlow DB: Increase in intra· cranial pressure during suclioni ng·stimulation vs, rise in PaC02, Anesthesiology 57:416-417, 1 982 Fox WW, Schwartz IG, Shaffer TH: Pulmonary physiotherapy in neonates: Physiological changes and respiratory management. Pediatrics 92:977981. 1978 Freedberg PO, Hoffman LA, Light WC, Kreps M: Ef· feet of progressive muscle relaxation on the ob· jective symptoms and subjective responses asso· dated with asthma. Heart Lung 1 6:24-30. 1 987 Fugl·Meyer AR: Effects of respiratory muscle paral. ysis in tetraplegic and paraplegic patients. Scand I Rehobil Med 3:141-150, 1971a Fugl.Meyer A R: A model for treatment of impaired ventilatory function in tetraplegic patients. Scond I Rehabil Med 3:168-177. 1971b Fugl·Meyer AR: Handbook ofCli ncal Neurology. /n­ juries of the Spine and Spinal Cord, Chap 19. The Respiratory System. pp. 335-349. American El­ sevier Publishing, New York. 1976 Fulton RL. lones CE: The course of post-traumatic pulmonary insufficiency in man. Surg Gynpcol 140:179-186. 1975 Garradd J . Bullock M: The effect of respiratory Iher· apy on intracranial pressure in ventilated neuro­ surgical patients. Aust J Physiol 32(2):107- 1 1 1 . 1986 Geisler FH. Salcman M: The head injury patient. In Emergency Su rgery and Critical Care. edited by JH Siegel. pp. 91 9-946. Churchill Livingstone. New York. 1987 Goldman 1M. Rose LS. Williams SI. Silver JR. Den­ ison OM: Effect of abdominal binders on breath­ ing in tetraplegic patients. Thorax 41 :940-945. 1 986 Gregory GA: Respiratory care of the child. Cril Care Med 8:582-586. 1 980 Guttmann L: Spinal Cord Injuries. Comprehensive Monagemenl and Research. 2nd ed. pp. 209-215. 575. Blackwell Scientific Publications. London. 1976 Haas A. Lowan EW. Bergofsky EH: Impairment of respiration afler spinal cord injury. Arch Phys Med Rehabil 46:399-405. 1 965 Haas F. Axen K. Pineda H. Gandino D. Haas A: Tem­ poral pulmonary function changes in cervical cord injury. Arch Phys Med Aehabil 66:1 39-144, 1986

Hammon WE. Kirmeyer PC. Connors AF. McCaffee DR, Kaplan RI: Effect of bronchial drainage on in­ tracranial pressure in acute neurological injuries (abs'rac'). Phys Ther 61 (5):735. 1981 Henke KG. Orenstein OM: Oxygen saturation dur­ ing exercise in cystic fibrosis. Am Rev Respir Dis

1 29:708-7 1 1 . 1 984 Hislop A. Reid L: Growth and development of the respiratory system-anatomical development. In ScientifiC Foundations of Pedialries. edited by IA Davis and , Dobbing. p. 221. WB Saunders. Phil­ adelphia. 1974 Holloway R. Adam EB. Desai SO. Thabiran AK: Ef· fect of chest physiotherapy on blood gases of ne· onates treated by intermittent positive pressure respiration. Thorax 24:421-426. 1 969 Holloway R. Desai MO. Kelly SD. Thambiran AK. Strydom SE. Adams EO: The effect of chest phys­ iotherapy on arterial oxygenation of neonates during treatmenl of tetanus by intermit ten posi­ tive pressure respiration. 5 Afr Med 1 40:445-447.

1 966

Huber AL. Eggleston PA. Morgan J: Effect of chest physiotherapy on asthmatic children (abstract). 1

Allergy Clin Immunol l l l :109- 1 10. 1974 Imle PC. Anderson PA. Ciesla ND: The effect of wearing an abdominal binder during the acute phase follOWing quadriplegia (abstract). Arch

Phys Med Rehabil 67:656. '986

Imle PC. Mars MP. Eppinghaus CEo Anderson P. Ciesla N: Effect of chest physiotherapy position­ ing on intracranial and cerebral perfusion pres­ sure (abstract). Crjj Core Med 16(4):449. 1988 !ennell B. Teasdale G: Management of Hp.od Inju­ ries. Davis. Philadelphia. 1981 Johnson ID. Malachowski NC. Grobstein R. et al.: Prognosis of children surviving with the aid of mechanical ventilation in the newborn period. J

Pedial 84:272. 1 974 Johnson TK. Moore EM. Jeffries I E editors: Children Arp Differen l: Developmenlal Physiology. 2nd ed. Ross Laboratories. Columbus. 1 978 Kattan M: Long term sequelae of respiratory ill ness in infancy and childhood. Pp.diulr Clin Norlh Am

26(3):525-535, 1979 Keens TG. Krastins IRB. Wannamaker EM. Livison H. Crozier ON. Bryan AC: Ventilatory muscle en­ durance training in normal subjects and patients with cystic fibrosis. Am Rev nespir Dis ' 1 6:853860. 1977 Kerrebijn KF. Veentzer R. Bonzet E. Water VD: The immediate effect of physiotherapy and aerosol treatment on pulmonary function in children with cystic fibrosis. Eur I Respi r Dis 63:35-42. 1 982 Klaus M, Fanaroff A. Marlin R: Respiratory prob­ lems. In Care of Ihe High Risk Neonale. edited by M Klaus and A Fanaroff. p. 1 90. Saunders. Phila­ delphia. 1 979 Landau L1: Management of a child with asthma Med J Ausl 1 :340-344. 1977 Ledsome JR. Sharp 1M: Pulmonary [unction in acute cervical cord injury. Am Rev Respir Dis 124:4144. 1981 Lerman RM. Weiss MS: Progressive resistive exor­ cise in weaning high quadri plegiCS from the ven­ tilator. Paraplegio 25:130-135. 1 987 Levine AL: Chest physical therapy for children with pneumonia. JAOA 78:122-125. 1978

279

CHEST PHYSIOTHERAPY FOR SPECIAL PATIENTS Livingstone JL: Physical trealment in asthma. Bf / Phys Med 1 5 : 1 36-139. 1952 Long fG. Philip AGS. Lucey JF: Excessive handling as a cause of hypoxemia. Pediotrics 65(2):203-207.

1 980 Lorin MI. Denning ER: Evaluation of postural drain· age by measurement of sputum volume and con­ sistency. Am J Phys Med 50:215-219, 1971 Lough MD. Doershuk CF. Stern RC: Pediatric Res­ piratory Therapy. p. 9. Year Book Medical Pub­ lishers. Chicago. 1974 Mackenzie CF. Shin S, Fisher R. Cowley RA: Two­ year mortality in 760 patients transported by heli­ copter di rect from the road accident scene. Am SlIrg 45:10'1-108, 1979 Macklem P1: Airway obstruction and collaleral ventilation. Am Rev Hespir Dis 1 1 6:287-289. 1977 Maloney FP: Pulmonary function in quadriplegia. Effects of a corsel. Arch Phys Med Rehabil 60:261265. 1979 March H: Appraisal of postural drainage for chronic obstructive pulmonary disease. Arch Phys Med Rehobil 1 1 :528-530. 1971 Martin RL. Herrell N. Rubin D, Fanaroll A: Effect of supine and prone positions on arterial oxygen tension in the preterm infant. Pediatrics 63(4):528-531. 1979 Mascia AV: Manual on the standardization of care of the severely asthmatic child. I Asthma Res 1 3 : 1 1 5-127. 1 976 Massery M: Respiratory rehabilitation secondary to neurological deficits: understanding the deficits. In Chest Physical Therapy and Pulmon a ry Reha· bilitation. edited by D Frownfelter, pp. 501, 541. 2nd ed. Year Book, Chicago, 1987 Matjasko I, Pilts L: Controversies in severe head in· jury management. In Clinical Controversies in Neurooneslhesia and Neurosurgery. edited by J Matjasko and J Katz. pp. 181-231 Grune & Strat· Ion. New York. 1 986 May DB. Munt PW: Physiologic effects of chest per· cussion and postural drainage in patients with stable chronic bronchitis. Chesl 75:29-32, 1979 McCool FD. Pichurko BM. Slutsky AS. Sarkarati 1\1. Rossier A. Brown R: Cha nges in lung volume and rib cage configuration with abdominal binding in quadriplegia. , Appl P hysioI 60(4): 1 1 98- 1 202. 1 986 . McKaba PG: Treatment of asthma in adults. Culis 17:1 1 1 5- 1 1 19. 1976 MCKinley CA, Auchincloss JH, Gilbert R. Nicholas J: Pulmonary function, ventilatory control. and respiratory complications in quadriplegic sub· jects. Am Rev Respir Dis 100:526-532. 1969 McMichan IC. Michel L, Westbrook PR: Pulmonary dysfunction following traumatic quadri plegia. lAMA 243:528-531. 1960 McQuillan KA: The effects of the trendelenberg po· sHion for postural drainage on cerebrovascular status in head·injured patients. Hearl Lung 16:327. 1 987 Menkes HA. Traystman RJ: Collateral ventilation. Am Rev Respir Dis 1 1 6:287-289, 1977 Metcalf VA: Vital capacity and glossopharyngeal breathing in traumatic quadriplegia. Phys Ther 46:835-838. 1 966 Montero IC. Feldman OJ, Montero 0: Effects of glos· sopharyngeal breathing on respiratory function after cen'ical cord transection. Arch Ph ys Med RehabiI 48:650-653. 1 967

Morraine I I . Brimioulle S, Kahn R: Active physical therapy and respiratory therapy effects on intra­ cranial pressure (Abstract). Crit Care Med

16(4):450. 1988 Mortola JP, Sant'Ambrogio G: Motion of the rib cage and the abdomen i n tetraplegic patients. Clin Sci Mol Med 54:25-32. 1 978 Motoyama EK: Lower airway obstruction. I n Fun­

damenlal Problems of Cystic Fibrosis ond Reloted Diseases. edited by JA Mangos and RC Talamo. pp 335-343. Stratton Intercontinental Medical Book Corporation. New York. 1973 Moulton A, Silver IR; Chest movements in patients with traumatic inju ries of the cervical cord. Cli n Sci 39:407-422. 1970 Muller NL. Bryan AC: Chest wall mechanics and respiratory muscles in infants. Pediotr Clin North Am 26(3):503-516. 1979 :Myers MG. Mcguiness GA, Lachenbrunch PA. Koontz FP, Holli ngshead R, Olson DB: Respira­ tory illness in survivors of infant respiratory dis· tress syndrome. Am Rev Respir Dis 133:101 1 1018. 1986 Newton DAG. Bevans HG: Physiotheray and inter­ mittent positive-pressure ventilation of chronic bronchitis. Br Med / 2:1525-1 528. 1978 Newton DAG. Stephenson A: Effect of physiother. apy on pulmonary function. Lancel 2:228-230, 1 978 Ok ken A, Rubin IL. Martin R I : Intermittent bag ventilation of preterm infants on continuous pos­ itive airway pressure. The effect on transcutane­ ous PO,. ) Peclialric 93(2):279-282. 1978 Oldenburg FA, Dolovich MB, Montgomery JM, Newhouse MT: Effects of postural drainage, ex­ ercise and cough on mucous clearance in chronic bronchitis. Am Rev Respir Dis 20:739-746. 1979 Orenstein OM. Franklin SA. Doershuk C . Heller· stein 11K, Germann KI. Horowitz IG. Stern RC: Exercise conditioning and cardiopulmonary fit­ ness in cystic fibrosis. Chest 80(4):392-397. 1981 Parker AE: Chest physiotherapy in the neonatal in­ tensive care unit. Physiotherapy 7 1 (2):63-65. 1985 Perlman J. Volpe J: Suctioning the preterm infant: Effects on cerebral blood now velocity. intracra­ nial pressure. and arterial blood pressure. Pedi· alrics 72(3):329-334. 1983 Petersen ES. Esmann V. Honcke P. Munkner C: Ef­ fect of treatment on chronic bronchitis. ACla Med Scand 182:295-303. 1 967 Pryor IA. Webber SA: An evaluation of the forced expiration technique as an adjunct to postural drainage. Physiotherapy 65:304-307. 1 979 Purohit OM. Caldwell C. Levkoff AH: Clinical memorandum, multiple rib fractures due to phys­ iotherapy in a neonate with hyaline membrane disease. Am I Dis Child 129:1103-1 104. 1975 Rava P. Yeh TF, Mora A. Pildes RS: Changes in transcutaneous P02 during tracheobronchial hy­ giene in neonales. Peri notology·Neono lology 4:41-45. 1 980 Reines AD, Harris RC: Pulmonary complications of acute spinal cord injuries. Neurosurgery 21(2): 1 93-1 96. 1987 Ries AL. Fedullo PF. Clausen JL: Rapid changes in arterial blood gas levels after exercise in pulmo­ nary patients. Chest 83(3):454-456. 1983 Ries AL, Moser K: Comparison of isocapnic hyper­ ventilation and walking exercise training at home

280

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

in pulmonary rehabilitation. Chest 90(2):285-289.

1 986 Roper PC. Vonwiller lB. Fisk ce. Gupta 1M: Lobar atelectasis after nasotracheal i ntubation in new­ born i nfants. Aust Poediofr J 1 2:272-275. 1976 Rossman eM, Waldes R. Sampson D. Newhouse MT: Effect of chest physiotherapy on the removal of mucus i n patients with cystic fibrosis. Am Rev Respir Dis 126:131-135. 1 982 Schumacker PT. Rhodes CR. Newell IC. Dullon RE. Shah DM. Scovill WA. Powers SR: Ventilation­ perfUSion imbalance after head trauma. Am Rev Respir Dis 1 1 9:33-43, 1979 Schultz H , Taylor FA: Intracranial pressure and ce­ rebral blood flow monitoring in head injuries. Can Med Assac J 1 1 6:609-613. 1977 ShaHt MN. Umansky F: Effect of routine bedside procedures on intracranial pressure. Israel J Med Sci 1 319):881-886. 1977 Siebens AA. Kirby NA. Poulos DA: Cough fol lowing transection of spinal cord at e-6. Arch Phys Med RehabiI 45:1-8. 1 964 Simbruner G . Coradello H, Fodor M, Havelec L. Lubec G. Pollak A: Effect of tracheal suction on oxygenation. circulation. and lung mechanics i n newborn infants. Arch Dis Child 56:326-330, 1981 Singh V: Effect of respiratory exercises on asthma­ the pink city lung exerciser. J Aslhma 24(6):355359. 1987 Speidel BD: Adverse effects of routine procedures on preterm infants. Lancet 1 :864-866. 1 978 Stein DA, Bradlwy BL. Miller WC: Mechanisms of oxygen effects on exercise i n patients with chronic obstructive pulmonary disease, Chest 8 1 :6-1 O. 1 982 Sutton pp, Pavia D, Bateman IRM. Clarke SW: Chest physiotherapy: A review. Eur J Respir Dis 63:188201 . 1 982 Sutton PP. Lopez-Vidriero MT. Pavia D, Newman SP. Cloy MM. Webber B. Parker A. Clarke SW: Assessment of percussion. vibratory shaking. and breathing exercises in chest physiotherapy. Eur J Respir Dis 66:147-152. 1 985 Stocks I, Godfrey 5: The role of artificial ventilation, oxygen and CPAP in the pathogenesis of lung damage in neonates, Assessment by serial mea­ surements of lung function, Pediatrics 57:352357. 1976 Streider 01: Pediatric origins of chronic obstructive lung disease. Bull Physiopolhol Respir 1 1 :273. 1 974 Tecklin IS. Holsclaw DS: Bronchial drainage i n pa­ tients with cystic fibrosis. Phys Ther 55:10811084. 1975 Tomney PM. Finer NN: A controlled evaluation of muscle relaxation i n ventilated neonates (ab­ stract). Cril Care Med 8:228. 1 980 Tudehope 01. Bagley C: Techniques of physiother­ apy in intubated babies with the respiratory dis­ tress syndrome. Aust Paedialr J 16:226-228, 1 980

Wailoo MP. Emery JL: Normal growth and devel­ opment of the trachea, Thorax 37:584-587, 1 982 Walker I . Cooney M: Improved respiratory function in quadriplegics after pulmonary therapy and arm ergometry. N Eng' J Med 3 1 6(8):485-487. 1987 Walsh CM, Bada H. Korones SB. Carter M, WongSP, Arheart K: Controlled supplemental oxygenation during tracheobronchial hygiene. Nurs Res

3614 ):211-215. 1987

Webb MSC. Martin lA, Cartl idge PH; NGYK. Wright NA: Chest physiotherapy in acute bronchiolitis. Arch Dis Child 60: 1078-1079. 1985 Webber BA: Current trends in the treatment of asthma. Physiotherapy 59:386-390. 1973 Weller PH. Bush E. Preece MA. Matthew 01: Short term effects of chest physiotherapy on pulmonary function in children with cystic fibrosis, Respi·

ralion 40:53-56. 1980 Wetzel J. Lunsford BR. Peterson MJ, Alvarez SE:_ Respiratory rehabilitation of the patient with R spinal cord injury. In Cardiopulmonary PhYSical Therapy. edited by S Irwin and J Tecklin. pp, 3954 1 1 . Mosby. St. Louis. 1 985 While PF. Schlobohm RM, Pitts LH. Lindauer JM: A randomized study of drugs for preventing in­ creases in intracranial pressure during endotra­ cheal suctioning. Anesthesiology 57:242-244,

1 982

Whitfield 1M. lanes MD: Atelectasis associated with mechanical ventilation for hyaline membrane disease. Cri! Core Med 8:719-732. 1980 Wicks AB. Menler RR: Long-term outlook in quad­ riplegic patients with initial ventilator depen­ dency. Chest 9013):406-410. 1 986 Wong JW. Keens TG, Wannamaker EM. Crozier ON, Levison H, Aspin N: Effects of gravity on tracheal mucus transport rates in normal subjects and in patients with cystic fibrosis. Pediatrics 60:146-

1 5 1 . 1977 Wood DW, Kravis LP. Lecks HI: Physical therapy for children with intractable asthma. J Asthma Res

7:1 77-182. 1970

Yeh TF. Leu ST. Pyati S. Pildes RS: Changes in O2 consumption in response to NICU care proce­ dures i n premature infants, Pedialr Res 16(2):315.

1 982 Zach M. Oberwalder B. Hansler F: Cystic fibrosis: Physical exercise versus chest physiotherapy. Arch Di, Child 57:587-589. 1 982 Zach M. Purrer e. Oberwaldner B: Effect of swim­ ming on forced expiration and sputurh clearance in cystic fibrosis, Lancet 1:1201-1 203. 1981 Zapletal A. Stefanova J. Horak J. Vavrova V. Sarna­ nek M: Chest physiotherapy and airway obstruc­ tion in patients with cystic fibrosis-a negative report. Eur J Respir Dis 64:426-433, 1 983 Zinman R. De Boeck C: Cough versus chest physio­ therapy: A comparison of the acute effects on pul­ monary function in patients with cystic fibrosis. Am Rev Rcspir Dis 1 29:182-184, 1984

CHAPTER 9

Adjuncts to Chest Physiotherapy P. Cristina Imle, M.S., P.T. flumidity Normal and Supplemental Humidity Controlled Environment Controlled Inspired Atmosphere IPPB Aerosol Delivery Work of Breathing Pulmonary Function Psychological E"ect Prevention of Pulmonary Complications Associated with Surgery Complications Associated with IPPB The Cost of IPPB Bronchodilating and Mucolytic Aerosols Bronchodilators Mucolytic Agents Complications Mechanical Devices Used to Encourage Lung Expansion Following Surgery Blow Bottles Incentive Spirometry Continuous Positive Airway Pressure and Positive Expiratory Pressure Bronchoscopy Complications and Precautions Restrictions Comparison with Chest Physiotherapy

Often the therapy given to patients re­ quiring intensive respiratory care affects both the airways and the types of secre­ tions they produce. As a result. there is much investigation into techniques for improving secretion clearance. Supple­ mental humidity is routinely given to most intensive care unit [ICU) patients; other methods advocated to clear pul­ monary secretions include intermittent positive pressure breathing [IPPB), aero­ sols. incentive spirometry [IS). continu­ ous positive airway pressure [CPAP). positive expiratory pressure [PEP). blow bottles. and bronchoscopy. This chapter discusses some of the i ndications. effec­ tiveness. and complications of these techniques. 281

HUMIDITY

Normal and Supplemental Humidity Adequate humidity is necessary for proper respiratory function. Studies on the humidification of inspired air show that by the time the gas reaches the sub­ glottic region of the trachea. it not only is warmed to 37'C but also is fully satu­ rated with water vapor [Robinson. 1974). It is well documented that ciliary activity is dependent on humidification levels [Dalhamn. 1956; Toremalm. 1961; Kil­ burn. 1967; Graff and Benson. 1 �f\9; As­ mundsson and Kilburn. 1970). Cilia ex­ tend from the respiratory bronchioles to the larynx [Hilding. 1 957). and ciliary ac-

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

282

tion is considered the most efficient and physiological means of cleansing the res­ piratory tract (Graff and Benson). There­ fore. it can be concluded that normal clearance of secretions from the lungs is also dependent upon proper humidifica­ tion. Both Dalhamn and Forbes (1973) found significant reductions or cessation of mucus flow at 50% relative humidity (RH)* levels in animals. Based on these findings. Graff and Ben­ son. along with many others. believe that all inspired gases must be humidified. if not by the nose and pharynx. then by ar­ t i ficial means-hence the use of humid­ ifiers. Various types of humidifiers are available. In the literature. addition of humidity is specifically recommended for patients who are intubated. venti­ lated. anesthetized. or receiving supple­ mental oxygen; for the newborn; and for those with severe chest injury. chronic obstructive pulmonary disease (COPD). asthma. pneumonia. atelectasis. respira­ tory burns. or innumerable other clinical conditions (Sara. 1965; Egan. 1967; Rashed et al.. 1 967; Chamney. 1969; Graff and Benson; Forbes; Downie. 1 9 79; and others). In short. it seems widely agreed that anyone breathing dry gases. having an artificial airway. or having abnormally thick secretions should receive supple­ mental humidity. There are two class­ ifications of humidifiers. controlled environment and controlled inspired atmosphere. Controlled Environment Controlled environment systems are applicable to spontaneously breathing patients but not necessarily those whose upper respiratory tracts are bypassed. They consist of such devices as fog rooms. steam or mist tents. and incuba­ tors. These systems are all constructed so that the patient is contained and cared for in the humidified environment. They suffer the same complication as conRelative humidity is the vapor content of a gas ex­ pressed as a percentage of that gas at full saluration at the same temperature. Another way of describing water content is in milligrams per liter. AI 31·C, fully saturated water content can be expressed as 44 mg/liter or 100% relative humidity. •

trolled inspired atmosphere systems in that they are susceptible to infection transmission. which is primarily bacte­ rial. Controlled environment devices are more expensive and limited in their mo­ bility. especially compared with most controlled inspired atmosphere equip­ ment (Chamney). Controlled Inspired Atmosphere Controlled inspired atmosphere sys­ tems alter the inspired gases of a patient but not the entire environment. They can be divided into four main groups. which are discussed in greater detail below: ( 1 ) heat a n d moisture exchangers o r con­ densers. (2) instillation or infusion meth­ ods. (3) nebulizers. both pneumatically driven and mechanically or ultrasoni­ cally activated. and (4) water bath humidifiers. Heat and Moisture Exchangers/ Condensers

Heat and moisture exchangers/con­ densers minimize heat and humidity loss from the upper respiratory tract and are commonly referred to as "artificial noses" (Walley. 1956; Toremalm; Maple­ son et al.. 1963; Siemens-Elema. 1979; Weeks and Ramsey. 1983). They function as follows: Humidified expired gases pass through a sponge. paper. metal. or gauze mesh. which causes condensation of moisture and heat retention. Most cur­ rently used condensers are composed of a synthetic felt and cellulose sponge. The retained heat and moisture are then added to the inspired gases. Heat and moisture exchangers can be used,during spontaneous or artificially controlled ventilation. Over the past decade. there has been increased emphasis on airway humidification during anesthesia as well as for postoperative management. When added to inhaled anesthetic systems. the newer condensers have been shown ef­ fective in conserving some of the heat and moisture loss that would otherwise occur. They are also easy to use and re­ quire no supplemental power source. The role of condensers for patients in the ICU or requiring mechanical venti­ lation for longer than 24 hr is more con-

ADJUNCTS TO CHEST PHYSIOTHERAPY

troversial. Primiano and associates (1 984) found that adding a condenser (compared with using no supplemental humidity source) improved the RH to 69.2% at body temperature in six ICU patients. Macintyre and co-workers (1983) re­ ported no significant difference in airway pressure, compliance, resistance, or ar­ terial blood gases in 26 ICU patients when co ,?ventional cascade humidifica­ tion was compared to a condenser for 24 hr. The authors also estimated sputum volume (over 4 hr) and radioaerosol clearance (over 1 hr) to be similar with both types of humidifiers. The tempera­ ture settings used during cascade humid­ ification in this study were not specified. Others question the use of condensers when longer periods of mechanical ven­ tilation are needed (Hay and Miller, 1982; Kahn, 1983; Perch and Realey, 1984; Cohen et aI., 1988). Significant increases in endotracheal tube occlusion (within 12 hr), pneumonia, atelectasis, and bron­ chial cast formation have been associated with condenser use compared with cas­ cade humidification (Perch and Realey; Cohen et al.). Microscopic studies on the effects of condenser humidification (less than 100% RH at body temperature) are limited. When inspired air is only 6070% saturated, there is evidence that the lower airways can supply additional heat and moisture for up to 3 hr. It is not "known what happens to the subcarinal airways of humans during prolonged per­ iods of reduced RH (Kahn) or reduced systemic hydration. Histologic damage to the tracheal epithelium of dogs exposed to desiccation and then rehumidification did not correspond with the noted changes in mucus velocity that rapidly improved (Hirsh et aI., 1975). Our clinical experience is similar to Cohen and asso-' ciates. We believe that condenser humid­ ifiers are not adequate for all ICU pa­ tients. Particular attention should be given to patients with thick secretions or respiratory muscle weakness and those who are spontaneously breathing, re­ quire delivered minute volumes of >10 liters/min, or a FlO, > 0.4. When con­ densers are used, they should be re­ placed with conventional humidifiers if sputum becomes tenacious or if difficulty occurs with suctioning.

283

Although there is some debate as to the minimal acceptable levels of supplemen­ tal humidity, the American National Standards Institute (1 979) suggests a min­ imum output of 30 mg H,O/liter gas while the Emergency Care Research In­ stitute (1983) recommends a minimum output of 21-24 mg H,O/liter gas. Not all condensers provide acceptable humidifi­ cation, particularly at increasing flow rates, tidal volume, or oxygenation. The overall effectiveness of condensers de­ pends on the heat and humidity·already present in the gas before exposure to the exchanger system. Mapleson et al. re­ ported that from 40 to 90% of the mois­ ture that might otherwise be lost may be retained by this method. Unlike the na­ sopharynx. which increases heat and moisture retention under colder and drier conditions, full saturation of dry gases is not possible with a condenser (Robinson, 1 9 74). The condenser's ability to humidify is inversely related to vol­ ume and FlO, delivery (Hay and Miller; Weeks and Ramsey; Perch and Realey; Cohen et al.). Heat and moisture ex­ changers also cause an increase in resist­ ance and dead space which should be considered during use with children and "borderline" patients during spontane­ ous breathing (Siemens-Elema; Weeks; Hay and Miller; Branson et aI., 1 986). In weak or critically ill patients the added breathing load imposed by the condenser may cause respiratory muscle fatigue and interfere with weaning from the ventila­ tor (Ploysonsang et aI., 1988). Cas leakage around the condenser is another concern, particularly with neonates; however, condenser humidification may be suc­ cessfully used in this patient population (Cedeon et aI., 1987). The earlier model condensers were found most effective when the gauze or mesh was kept reasonably dry. Frequent changing was thought to minim ize bac­ terial contamination. There is some evi­ dence that the new type of heat and moisture exchangers may reduce venti­ lator contamination by trapping exhaled bacteria in the inner core of the humidi­ fier (Stange and Bygdeman, 1980). This has not been associated with an in­ creased risk of airborne inhaled bacteria during mechanical ventilation (Powner

284

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

et al.. 1986). However. when mucus is trapped in the condenser. the system should be changed due to the increased airway resistance or obstruction (Robin­ son; Sykes et al.. 1976; Weeks); other­ wise. changing the heat and moisture ex­ changer every 24 hr is recommended (Siemens-Elema). InstillBtion/lnfusion

Instillation/infusion is a simple yet controversial method of humidifying in­ spired gases. It consists of instilling a fluid. usually physiological saline or a mucolytic agent. at a set rate into the in­ spiratory limb of the ventilator tubing or directly into the tracheal tube. Though this method is easy to perform. the effect is probably no different from instilling fluid directly into the trachea (see p. 180). Hayes and Robinson (1970) found that humidity levels decreased sharply when this method of humidification was used in comparison to a hot water bath humid­ ifier or a nebulizer. Instillation can also lead to increased airway resistance (Sykes et al.). Another hazard of instilla­ tion is the uncontrolled or improperly controlled rate of infusion. which could lead to drowning. At best. this intermit­ tent flooding of the respiratory mucosa is unlikely to be an acceptable physiologi­ cal substitute for normal conditions and may have serious consequences. espe­ cially if large amounts of fluid are ab­ sorbed through the lungs (Huber and Fin­ ley. 1965; Chamney). It is doubtful that instillation of fl uid into the artificial air­ way has any effect on humidifying distal airways. Although it may be helpful in compensating for humidity loss from the upper airways of an intubated patient. it is considered an u nsatisfactory method of humidification during mechanical or spontaneous respiration (Hayes and Robinson). Supplemental saline or water is often added during high frequency ventilation (HFV) where adequate airway humidifi­ cation has emerged as a problem (Berman et al.. 1984; Ophoven et al.. 1 984). Two gas sources. entrained and injected. must be considered during this mode of venti­ lation. Moisture may be added to en-

trained gases with a nebulizer or water bath humidifier. while some method of instillation/infusion is commonly used to humidify the injected gas (Berman et al.. Doyle et al.. 1984; Ophoven et al.). For paitents with adequate lung compliance and no significant lung pathology. the ratio of entrained-to-injected gas is high. Therefore. entrained gas that is optimally humidified may partially compensate for the effect of the dry injected gas. Prob­ lems with tracheal m ucosa damage arise when the proportion of entrained gas de­ creases. as occurs in patients with pul­ monary pathology or poor compliance. who may require HFV. When instilla­ tion/infusion of injected gas is used without humidified entrained gas. mucus transport is markedly reduced (Klain et al.. 1982). Warming the instillate or the injected gas does not prevent heat loss since rapid cooling occurs during the aerosolization of the infused water or sa­ line. These factors are thought to be par­ tially responsible for the increase in in­ spissated secretions and tracheal damage seen with HFV (Berman et al.; Doyle et al.; Ophoven et al.). Nebulizers

There are two basic types of nebuliz­ ers. the pneumatically powered and the mechanically or ultrasonically activated. Both can be used on spontaneously breathing patients as well as those re­ quiring intubation or mechanical venti­ lation. However. nebulizers are reported to interfere with mechanical ventilation (Sara and Clifton. 1962; Glover. 1965; Bo­ somworth and Spencer. 1965; Hayes and Robinson; Klein et al.. 1973). If the gas flow delivered to the patient by the ven­ tilator also operates the nebuli ter. inad­ equate humidity can result. especially at low flow rates (Hayes and Robinson; Klein et al.). Alternatively. if the nebu­ lizer is driven by an auxiliary oxygen supply. humidification may improve. but the concentration of delivered oxygen can also increase significantly (Bosom­ worth and Spencer). Pneumatically Powered Nebulizer. The pneumatic-powered humidifier works on the Bernoulli prinCiple. A narrow jet of

ADJUNCTS TO CHEST PHYSIOTHERAPY

high-pressure gas is blown across a water reservoir. Both water vapor and droplets are formed and blown against an anvil or system of bames, which usually filters out the larger water particles. The gas en­ tering the patient's delivery tube still contains a visible mist of both water vapor an� droplets ranging from 5 to 20 II­ (Robinson). This dense mist does not nec­ essarily imply complete gas saturation with humidity, since visible mists with relative humidities of less than 78% are possible (Robinson). Klein and associates found that no correlation could be made between the visible mist and the water content of a gas. Heated and unheated pneumatic-powered humidifiers are available commercially. Studies show that humidities of only up to 29 mg/liter (at the patient's end of the delivery tube) are possible with unheated nebulizers, while up to 50 mg/liter can be achieved using heated types (Sara and Curie, 1965; Sara; Bosomworth and Spencer). More re­ cently, some unheated pneumatically driven nebulizers are reported to deliver nearly 100% relative humidity (Klein et al. ). Marked condensation along the deliv­ ery tube is a problem, particularly with heated pneumatic-powered humidifiers. In fact, this system should not be used with narrow-gauge delivery tubing, since condensation can effectively block both gas and humidity flow to the patient. jones and co-workers (1 969) demon­ strated how a fluid-lined tube could in­ crease resistance to gas flow. Much as a fluid-lined delivery tube can increase re­ sistance or block air flow, the inhalation of high-density mists can i ncrease the fluid lining of the respiratory tract, caus­ ing greater airway resistance (Robinson). Due to the small water particle size de­ livered to the airways, the problem of transmitting bacterial i nfections is also great. Therefore, cleaning and steriliza­ tion are extremely important. Despite these complications, the pneumatically driven humidifiers are more effective than those previously described, espe­ Cially when used on the spontaneously breathing patient. Mechanically or Ultrasonically Acti­ vated Nebulizer, The small particle size

285

of ultrasonically or mechanically acti­ vated nebulizers is achieved by dropping water onto a disc rotating at high speeds or onto a crystal Vibrating at very high frequencies (Chamney). These methods are reported to produce varying levels of humidification ranging from 30 to 200 mg/liter (Smith, 1966; Robinson). The potentially large amount of humidity de­ livered by this method, though thought to be effective in loosening thick secre­ tions by acting as a solvent, has also led to much criticism (Robinson; Downie). Because of the lack of output control on ultrasonic nebulizers, fluid gain, in­ creased airway resistance and water in­ toxication were reported to occur, espe­ cially in i nfants (Glover; Harris and Riley, 1967; Modell et aI., 1967; Cheney and Butler, 1968; Pflug et aI., 1970; Malik and jenkins, 1 9 72). Modell and associates studied the effects of continuous ultra­ sonic nebulization on dogs. Pathological changes compatible with bronchopneu­ monia were seen i n all eight dogs ex­ posed to a physiological saline aerosol. This effect was only seen in two of the eight dogs receiving nebulized distilled water, yet five of the eight demonstrated mild to moderate focal atelectasis. Other authors i nvestigating the effects of short-term use of ultrasonic nebuliza­ tion reported excessive coughing and wheezing in patients with chronic l u ng disease a nd in normal subjects (Cheney and Butler; Pflug et al; Malik and jen­ k i ns). Cheney and Butler found in­ creased airway resistance in patients and normal subjects using ultrasonic nebuliz­ ers. This repsonse was not seen with other forms of nebulizers. In patients with chronic bronchitis, Pflug and co­ workers reported significant decreases i n forced expiratory volume i n 1 sec (FEV,) and vital capacity following ultrasoni­ cally delivered aerosols. Similarly, in­ creased airway resistance was noted by Malik and jenkins, who reported that these physiological alterations were most marked when a 5% saline solution was nebulized, as opposed to distilled water or normal saline. The results of these studies provide evidence that a seem­ i ngly i nnocuous procedure, such as ul­ t rasonic nebulization of water or saline,

286

CHEST PHYSIOTHERAPY I N THE INTENSIVE CARE UNIT

produces definite side effects that may be disastrous, particularly i n patients with chronic airway disease (Cheney, 1 972). Ultrasonic nebulization can deliver water particles of 1-5 JL in size. In fact, 97% of ultrasonically nebulized humid­ ity, but only 55% of the pneumatically nebulized humidity, are delivered in par­ ticles within this range (Moffet et aI., 1967). A droplet-size spectrum of 2-10 JL is recommended for water deposition throughout the airways, since alveolar sacs and ducts may be theoretically reached by 1 - to 3-JL droplets, bronchioles by around 6-JL droplets, and bronchi and upper airways, by 10-JL droplets (Robin­ son). However, Sawyer (1963) states that in the spontaneously breathing patient, approximately 50% of particles 1-5 JL i n size are retained in the nasopharynx, while only 1 0-40% are deposited in the depths of the lungs. The pharynx and upper trachea are bypassed in tracheally intubated patients. As a result, droplet delivery to the smaller airways may be altered. Aerosol deposition is dependent on many variables and is not well understood. The complication of transmitted infec­ lion is a greater hazard with ultrasonic and mechanical nebulization due to the higher percentage of minute-sized water particles. The smaller size potentially al­ lows particles to reach the distal airways where pulmonary clearance mechanisms may not be as efficient. This is com­ pounded by the fact that these same par­ ticles can carry a signficant number of vi­ able bacteria (Reinarz et aI., 1965; Edmondson et aI., 1966; Ringrose et aI., 1968). The small droplet size also allows some particles to leave the lungs during expiration and to remain airborne, thereby providing a vehicle of transmis­ sion from one patient to another. Generally, pneumatic and ultrasonic nebulizers have four disadvantages: (1 ) There is poor control over the upper level of humidity delivered; (2) delivery tubes require frequent checking to pre­ vent obstruction by condensation; (3) lengthy warm-up periods may be neces­ sary to achieve steady outputs in some heated models; and (4) nebulizers can in­ troduce massive doses of bacteria to the

respiratory tract. This has a potentially great impact because bacteria carried by small water particles may be delivered to the distal airways (Chamney; Klein et al.; Sykes et al.). WBterBBth

In water bath humidifiers, inspired gas is either blown over or bubbled through water. This may allow full saturation for a given temperature. The humidified gas then passes through a delivery tube to the patient. Both heated and unheated water bath models are available. The un­ heated models are capable of saturating a gas only at ambient temperature and, therefore, do not allow full saturation at body temperature. Consequently, only inadequate humidities of 7-22 mg/liter are possible (Wells et aI., 1963; Bosom­ worth and Spencer; Darin et aI., 1982). Heated water bath humidifiers lead to heavy condensation in the delivery tubes. This results from heating and sat­ urating the air at above body tempera­ lure, therefore increasing the water vapor-carrying capacity, and then allow­ ing the air to cool while in the delivery tubes. Heated models can provide up to 42 mg/liter at 35'C (Wells et al.; Cham­ ney). Humidifiers with a thermostat on the inspiratory limb of the ventilator tub­ ing can deliver up to 44 mg H20/liter al 37'C (Robinson). Because of the amount of condensa­ tion, water bath humidifiers should al­ ways be kept lower than the patient to prevent accidental spillage into the pa­ tient's airway. Excess condensation also often calls for frequent delivery lube drainage or moisture traps. If the temper­ ature at the patient's end of the delivery tube is controlled, condensation can be minimized (Chamney). Problems of bac­ terial infection from waler bath, humidi­ fiers are reported low. nearing that of am­ bient air (Reinarz et al.; Edmondson el al.; Moffet et a I . , 1967; Schulze et aI., 1967). This is possible because water bath systems deliver humidity in vapor form, which does not carry bacteria (Chamney). Although condensate is fre­ quently contaminated during 24 hr of mechanical ventilation, the patient's se-

ADJUNCTS TO CHEST PHYSIOTHERAPY

cretions are reported as the primary source of colonization. Nonetheless, care should be taken to prevent inadvertent lavage with the condensate (Craven et aI., 1984). Also, water bath humidifiers can be used on both ventilated and spon­ taneously breathing patients, since their humidit¥ production tends to be less sus­ ceptible to changes in flow or tidal vol­ ume (Hayes and Robinson). Summary There is a wide range in the effective­ ness of commercially available humidifi­ ers. The optimal amount of water content that should be added to inspired gases is still a matter of controversy. It is gener­ ally agreed that humidification should approach 44 mg H20/liter. The three ac­ ceptable methods of humidifying gases use a heat and moisture exchanger or condenser, a nebulizer, or a water bath system (Robinson). Instillation of infu­ sion is of no proven benefit except possi­ bly with HFV, where appropriate airway humidification remains a problem. The new types of condensers are able to re­ tain most of the heat and moisture that would otherwise be lost, but they are less effective when increased oxygenation or volume delivery is necessary. Condens­ ers increase resistance and their use with pafients requiring mechanical ventila­ tion for longer than 24 hr remains under scrutiny. When used, condensers should be replaced with conventional humidifi­ ers if complications occur with inspis­ sated secretions or during spontaneous ventilation. Both water baths and nebulizers cause condensation in the delivery tubing; they should be frequently checked to prevent blockage or inadvertent patient lavage. By controlling the temperature of the pa­ tient's inspired air, condensation pro­ duced by water bath humidifiers can be markedly decreased. Ideally, all heated humidifiers should contain a tempera­ ture control alarm to warn of potential overheating. Though both types of neb­ ulizers may provide adequate humidifi­ cation, in comparing the two, the heated, pneumatically driven type is preferable. It is up to one-sixth the cost, and its drop-

287

let size is theoretically of greater benefit than that of the ultrasonic nebulizer (Robinson). There are fewer complica­ tions of overhumidification and in­ creased airway resistance associated with pneumatic nebulizers. Both meth­ ods of nebulization interfere with me­ chanical ventilation. However, heated water bath humidifiers are surprisingly efficient and, when incorporated in a ventilator circuit, create less interference with ventilation (Chamney; Hayes and Robinson; Robinson). They probably rep­ resent the best method of providing hu­ midity for the majority of patients (Sykes et a 1 . ). Water bath humidifiers are also less prone to bacterial contamination than the nebulizers. Although often re­ ferred to as "old fashioned," water bath humidi fiers are recommended for the pa­ tient in the ICU who requires mechanical ventilation for more than 24 hr. As with all types of heated humidifiers, care must be taken to prevent overheating. IPPB

The effects of IPPB have been studied since its introduction over 40 years ago (Motley et aI., 1948). During this time, it has been prescribed for use both before and after surgery and in the treatment of COPD. asthma, emphysema, and cystic fibrosis. Because of this wide spectrum, it is hard to compare the results found in one group of patients with those in an­ other. Much of the original information on IPPB was based on studies of patients with chronic lung disease. These findings were often extrapolated and claimed to be valid for surgical patients with acute secretion retention. The most commonly cited benefits of lPPB include improved aerosol delivery, reduced work of breath­ ing, improved respiratory function in pa­ tients with chronically diseased lungs, psychological support, and prevention of pulmonary complications in the surgical patient. Currently, there is little scien­ tific support for IPPB as a therapeutic mo­ dality (Gold, 1982; American Thoracic Society; 1987). This section addresses the claimed benefits and complications of IPPB, including its uncontrolled growth and cost.

288

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Aerosol Delivery IPPB is commonly used to deliver aer­ osols. There is little controversy in the literature about its efficacy in this area. Many authors report that IPBB is useful in delivering bronchodilating aerosols, though IPPB is not known to have any bronchodilatory effect of its own (Gold­ berg and Cherniack, 1965; Chang and Levison, 1972; Smeltzer and Barnett, 1973; Cherniack, 1974; Loren et aI., 1977; Sackner, 1978). In fact, significant in­ creases in airway resistance, air trapping. and dead space, as well as reductions in expiratory flow rates and alveolar venti­ lation, were found following IPPB when no bronchodilator was used (Wu et a I . , 1955; Kamat et a I . , 1962; Moore e t a I . , 1972; Ziment, 1973; Fouts and Brashear, 1976). Increased sputum production without correspond i ng improvement in vital capacity or FEY, was also reported following IPPB, implying that mucus pro­ duction may be stimulated by IPPB itself (Shim et aI., 1978). The real question is not whether IPPB is effective in del ivering aerosols but, rather, whether IPPB is the most effective method of delivery. The research done in this field is extensive, though Petty (1974) states that no well-designed study has shown convincingly that a broncho­ dilator, or any other drug delivered by IPPB, is more effective than inhalation of the same aerosol delivered by a powered or hand-held nebulizer. N umerous stud­ ies support Petty's claims (Froeb, 1960; Goldberg and Cherniack; Smeltzer and Barnett; Cherniack; Cherniack a nd Svan­ hill, 1976; Loren et al.; Shim et al.). Chang and Levison found no difference in deliv­ ering isoproterenol by means of IPPB or a powered nebulizer, except that the latter produces the same results at much lower dosages (5 mg compared to 0.225 mg). Goldberg and Cherniack and Wu and as­ sociates suggest that IPPB may be of added benefit for the patient who is un­ able to take a deep breath or coordinate breathing with a hand-held nebulizer. However, Murray (1974) states that the inability to teach patients how to use a hand-held or compressor-driven nebu­ lizer constitutes a failure of instruction, not an indication for IPPB. If a patient is

able to breathe, a simple aerosol genera­ tor can be used just as effectively as IPPB and possibly better (0' Donohue, 1982), Aerosol deposition to the larger or smaller airways is not enhanced by IPPB compared with quiet breathing (New­ house and Ruffin, 1978). Ziment (1973) delineates four disad­ vantages of aerosol delivery with IPPB: (1) Because gas flow follows the pathway of least resistance, aerosols are preferen­ tially delivered to the compliant airways. Therefore, less is delivered to the more spastic airways that may benefit most from bronchodilator therapy. (2) Con­ trolled dosages of medication delivery are not possible with IPPB. Aerosol dos­ ages may be 10-20 times greater than those used subcutaneously or intrave­ nously. yet only 5-15% of a total dosage is normally retained in the lungs. If, how­ ever, IPPB is connected directly to an ar­ tificial airway and no dosage reduction fs made, considerably larger quantities may be retained in the body. Systemic effects are not eliminated with aerosol delivery, since variable amounts of the medication may be retained in the mouth, esopha­ gus, stomach, duodenum, and trachea as well as the lungs. (3) Some inhaled bron­ chodilators that are beta stimulators, such as isoproterenol, result in pulmo­ nary vascular vasodilation. This can cause increased ventilation/perfusion mismatch, particularly in the presence of severe bronchospasm (Pierson and Grieco, 1969). (4) IPPB is more expensive yet no more effective in delivering bron­ chodilators and m ucolytic agents than are hand-held or compressor-driven ne­ bulizers. Water or humidification is be­ lieved to be the best mucus-softening agent and is without the side effects of currently used medications. Work of Breathing The ability of IPPB to affect the work of breathing has caused much debate in the literature. Sukuma1chantra and co-work­ ers (1965) found that IPPB increased the work of breathing in some patients but reduced it in others. Sinha and Bergofsky (1972) reported that IPPB at large infla­ tion pressures (average, 22 cm H,O) may be beneficial in patients with kyphosco-

ADJUNCTS TO CHEST PHYSIOTHERAPY

Iiosis and chronically decreased lung compliance. In six patients this parame­ ter was found to increase 70%, for up to 3 hr, and was accompanied by decreases in the work of breathing. Recent studies on patients with neuromuscular disease do nol sub&lantiale these findings. Pa­ tients with muscular dystrophy and quadriplegia demonstrated no improve­ ment in respiratory system compliance (neither chest wall nor static lung com­ pliance) following IPPB treatment that delivered volumes up to three times rest­ ing tidal volumes (DeTroyer and Deisser, 1981; McCool et aI., 1986). In part, the ability of IPPB to alter the work of breathing appears to depend on the degree of patient cooperation and re­ laxation (Ayres et al.. 1963; Sukumal­ chantra et al.). Bader and Bader (1969) noted that increased respiratory work was associated with high flow rates. Sim­ ilar findings were reported by Ayres and associates, when patients actively led the IPPB apparatus. Increases in airway col­ lapse, expiratory resistance, and air trap­ ping, which are associated with IPPB, may also result in a rise in the work of breathing Uones et aI., 1960; Kamat et al.; SukumaIchantra et al.; O'Donohue, 1982). Alterations in the work of breath­ ing, that may accompany IPPB, were not found to change the overall metabolic rate. The fraction of total body metabo­ lism expended on ventilation is normally small. As a result, if a decrease in respi­ ratory work is accompanied by i ncreased non respiratory work, such as agitation or discomfort (also attributed 10 IPPB), the desired effect may be neutralized (Suku­ malchantra et al.). The data generated on the work of breathing associated with IPPB must also be viewed in light of the difficulty in making accurale and repro­ ducible measurements of this parameter, especially in spontaneously breathing patients. Pulmonary Function Tidal Volume

IPPB is often prescribed because it is felt to increase tidal volume and, there­ fore, prevent small airway collapse. Cul­ len and co-workers (1957) concluded that

289

IPPB did cause an increase in tidal vol­ ume in normal subjects but not in pa­ tients with emphysema. A study on similar patients reported significant in­ creases in tidal volume with mechanical chest percussion, IPPB, manual chest percussion and voluntary deep breathing (Petty and Guthrie, 1971). Other re­ searchers claimed that the effects of IPPB on gas distribution and tidal volume were similar to those of voluntary hyper­ ventilation (Torres et aI., 1960; Emman­ uel el aI., 1966; Wohl, 1968; McConnell et aI., 1974). For persons who are unable to increase Iheir tidal volume voluntarily, Torres and associates suggested IPPB as a way of improving pulmonary gas distribution. Petty and Gu thrie noted that increased dead space ventilation was associated with deep breathing but not with IPPB. Other researchers reported the opposite. SukumaIchantra and co-workers found that the majority (mean, 56%) of the tidal volume increase achieved by IPPB re­ sulted in increased dead space. Volun­ tary hyperinflation by the same patients caused marked improvement in alveolar ventilation, and only a small portion (mean, 17%) of the tidal volume increase was dead space. This suggests that IPPB, in comparison to voluntary hyperinfla­ tion, overventilates alveoli that are al­ ready well ventilated. Consequently, its effect is wasted. In order to improve al­ veolar ventilation, increased driving pressures were suggested. However, this can cause increased alveolar pressure that may result in pulmonary vascula­ ture compression (Riley, 1962; Daly et aI., 1963). Therefore, increasing inspiratory flow or pressure may decrease perfusion, creating an even greater ventilation/per­ fusion mismatch. In eight supine subjects, Bynum and associates (1976) studied the effects of IPPB and spontaneous breathing at tidal volume and large lung volumes (greater than twice the tidal volume). The effect of these maneuvers on the distribution of ventilation, perfusion, and ventilation/ perfusion ratios was measured by using radioactive gas techniques. During tidal volume breathing, ventilation and per­ fusion were diminished in the lung bases (areas adjacent to the diaphragm) as com-

290

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

pared to other areas. However. this de­ crease was greater in the subjects receiv­ ing IPPB. despite the fact that the volume of air. inspiratory flow. and frequency of breathing were the same for both groups. Both voluntary and IPPB-induced hyper­ inflation caused improved basilar venti­ lation and perfusion; yet. the spontane­ ously increased volume resulted in significantly higher ventilation/perfu­ sion ratios than were found with IPPB. Bynum et al. concluded that at similar lung volumes. IPPB was inferior to spon­ taneous breathing in ventilating and per­ fUSing the lung bases. As a result. vol­ untary hyperinflations may be more ef­ fective in preventing atelectasis than lPPB. IPPB should not be considered as an alternative to deep breathing unless an increase in volume of at least 25% of that obtained by voluntary deep breath­ ing is obtained. It is generally accepted that IPPB has no benefit over volitional hyperinflation. in part because it empha­ sizes the least appropriate component of inspiration. pressure rather than volume (George and O·Donohue. 1980 and 1982; Gold. 1982; American Thoracic Society. 1987).

ArterialBlood Gases

Blood gas changes that occur after IPPB treatment were found both to improve and to worsen. Following IPPB. reduced arterial carbon dioxide levels were re­ ported in some patients and normal sub­ jects (Cullen et al.; Sukumalchantra et al.; Petty and Guthrie). However. signifi­ cant decreases in PaCO, were also dem­ onstrated following mechanical and manual chest compression on patients with obstructive lung disease (Petty and Guthrie). The reported changes in arte­ rial blood gases following IPPB were of short duration. Ziment (1973) empha­ sized that these fluctuations may be harmful. since short-lived decreases in PaCO, or increases in PaO, following IPPB may diminish respiratory drive and result in hypoventilation. Reduced res­ piratory rates were also noted by Cullen and associates (1957). In general. it appears that IPPB may in­ crease tidal volume. However. the same

or more beneficial effects are usually seen with voluntary hyperventilation. When IPPB does increase tidal volume. it is not clear in the literature if this corre­ lates with either increased alveolar ven­ tilation or improved arterial blood gases. None of these changes is reported to be of any long-term significance (Bader and Bader; Morris et al.. 1970; Ziment. 1973; Leith. 1974; Schemer and Delaney. 1981). Murray concludes that neither hypox­ emia nor CO2 retention is an indication for IPPB. since any change is only tran­ sient. Instead. significant CO, retention. hypoxemia. and acute ventilatory failure are more often indications for mechani­ cal ventilation. Maneuvers that attempt to treat the symptoms and provide tran­ sient improvement in gas exchange are doomed to certain failure unless the pre­ cipitating cause of respiratory failure is removed. Psychological Effect Whether IPPB is of psychological ben­ efit is controversial. The claim by the pa­ tient of "I feel better" after receiving IPPB should not be ignored. This. in it­ self. may be an important finding. despite the fact that laboratory tests failed to sub­ stantiate any measurable changes (Mur­ ray; Thornton et al.. 1974). Particularly in the capo patient. improvement in activ­ ities of daily living or quality of life fol­ lowing intensive rehabilitation are doc­ umented. while no significant changes in pulmonary function tests or correspond­ ing signs of disease reversal may be noted. However. subjective claims of im­ provement following IPPB treatment may be unreliable for the following reasons: (1) Some patients may not wish to disap­ point their physician or therapist and therefore. claim improvement. (2) Tran­ sitory improvement of symptoms is com­ mon when initiating a new treatment program. especially in patients with chronic diseases. Improvement is also noted when the person performing the treatment is highly motivated or enthu­ siastic. (3) The mystique surrounding an expensive and complicated-looking ma­ chine that makes hissing noises and emits clouds of vapor can undoubtedly serve to persuade some patients that dra-

ADJUNCTS TO CHEST PHYSIOTHERAPY

matic relief is imminent (Murray). The problem of psychological dependence, with LPPB serving as a security blanket, is most often. described in reference to long-term users, such as COPD patients. Murray states that the machine pre­ scribed to alleviate symptoms of pulmo­ nary disease has developed into a com­ plication that is more debilitating than the disease for which it was prescribed. There is no evidence that IPPB is helpful or desirable for home use (IPPB trial group, 1983). Prevention 01 Pulmonary Complications Associated with Surgery IPPB is believed by some to prevent, as well as treat, atelectasis after surgery by dilating collapsed bronchi and expanding underventilated atelectatic alveoli. To quote McConnell and his associates, it seems that "since Elisha (11 Kings 4:34) first used positive pressure breathing to resuscitate a Shunammite child, this method has found great favor with the medical profession." Because pulmonary complications account for 6-70% of prob­ lems seen following upper abdominal surgery (Pontoppidan, 1980), it is no won­ der that all posssible methods of redUCing this mtlrbidity were tried. The over­ whelming acceptance of IPPB, as judged by its widespread use, would lead one to believe it is beneficial; yet, the routine use of IPPB prior to su rgery is of un­ proven value in preventing atelectasis following surgery (Ziment, 1974; Gold, 1982). Similarly, its efficacy in preventing complications, such as atelectasis and pneumonia following surgery, has not been proven in any study of acceptable design (Petty). IPPB is most commonly administered with the patient in the head-up position. This may aid the flow of secretions to the dependent lung zones. Postoperative pul­ monary complications most commonly occur in the lower lobes of adult patients (Jaworski et al" 1988; Appendix 1.3). Therefore, the sitting postion is the most unfavorable for postural drainage in this patient population. IPPB also delivers an inspiratory pressure that may impede normal mucus flow. Based on these find­ ings, Ziment (1973) feels that physiother-

291

apy (including postural drainage, percus­ sion and supervised coughing) should be given after IPPB or the patient may well be harmed by the IPPB treatment. In 15 patients following surgery, Jones (1968) found that voluntary deep breath­ ing was more effective in i ncreasing tidal volume than IPPB at conventional pres­ sures (15 cm H20). When pressures of 25 cm H20 were used, IPPB was superior. It must be emphasized that this study did not attempt to establish whether IPPB at 25 cm H20 was beneficial in preventing complications following s urgery (Ziment, 1974). It was merely concerned with methods to improve tidal volume. Se­ quencing of the various therapies studied was not altered, and rest periods of only 10 min existed between trials. Conse­ quently, there may have been an additive effect, since the values may not have re­ turned to normal between therapies. Also, no follow-up was documented as to the long-term effect of these maneuvers in preventing airway collapse. In contrast, O'Donohue (1979) reported that IPPB was effective in the manage­ ment of pulmonary atelectasis in four case reports (only one involved a surgical patient). Changes in arterial blood gases and chest x-rays were the criteria for improvement. Inspiratory pressures be­ tween 35 and 45 cm H20 were used. These values are in excess of the normal pressures prescribed during IPPB, yet complications of mediastinal emphysema (reported following conventional pres­ sures) were not noted. Although chest physiotherapy, incentive spirometry, broncho-dilators, antibiotics, and fiber­ optic bronchoscopy were initially found ineffective in clearing the atelectasis in these patients, some of these treatments were continued along with maximal vol­ ume IPPB therapy. IPPB treatments were performed every 2 hr and repeated for 36-72 hr. Because the degree of patient mobilization was not specified. it is pos­ sible that some of the pulmonary im­ provement was due to increased patient activity (see Case Histories 6.1 and 6.2). Similarly, since other therapies were given simultaneously, it is difficult to conclude that any changes in atelectasis or arterial blood gases were due to the ef­ fect of IPPB alone.

292

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

Meyers et al. (1975) studied the effects of IPPB on functional residual capacity (FRC) in 10 patients following abdominal surgery. This parameter was chosen be­ cause it was believed to most accurately reflect alterations in alveolar ventilation or collapse. IPPB was found to have no influence on changes in FRC. Browner and Powers (1975) also investigated the effects of IPPB on FRC in patients follow­ ing surgery. Of those having normal or reduced FRC prior to treatment, they found that IPPB caused a significant fall (416 m l ) in this value. They also reported significant decreases in PaD, and oxygen saturation. Therefore, they suggest that administering IPPB to patients after sur­ gery may not only be of questionable benefit but may be potentially harmful. N umerous studies were undertaken to evaluate IPPB and its effectiveness in preventing or treating pulmonary com­ plications after surgery. Baxter and Lev­ ine (1969) studied 200 patients. Compar­ ing radiological and clinical findings before and after surgery, they concluded that IPPB was not effective in reducing the incidence of pulmonary complica­ tions. Becker et al. (1960) studied 100 pa­ tients after upper abdominal surgery and concluded that routine IPPB (2-3 times/ day for 3 days) did not prevent or clear atelectasis, when compared with a con­ trol group. Cottrell and Siker (1973) reached the same conclusion in studying 60 patients treated before and after sur­ gery. Although pulmonary function tests were found to improve in the patients with COPD who received IPPB before their operation, findings following sur­ gery were not affected. In evaluating the effectiveness of IPPB for patients following thoracic surgery, McConnell and colleagues concluded that although the depth of respiration was increased with IPPB, verbal encour­ agement of the patient to breathe deeply was just as effective. Sands and co-work­ ers (1961) studied 84 patients receiving IPPB after upper abdominal surgery and found results in agreement with those described above. They also noted that the patients treated with IPPB complained of "more mucus" than did the control group, but that this did not lead to in­ creased cough stimulation or subsequent

expectoration. In addition, IPPB was not found to reduce discomfort after surgery, to result in earlier patient ambulation, or to result in decreased hospital stay, as was expected. On the other hand, Anderson et al. (1963) found that pulmonary complica­ tions were markedly decreased when IPPB was used following surgery. Baxter and Levine commented on this study as involving a far more heterogeneous group of patients and attributed the low pulmonary complication rate to the dili­ gence of the therapist rather than the IPPB apparatus. Fouts and Brashear state that Anderson and co-workers' conclu­ sions are difficult to accept because of the disparity in nu mbers (160 control pa­ tients and 42 treated), and the lack of in­ formation about the respiratory care given to the controls. Noehren and asso­ ciates (1958) may have a better explanation: , Most suggestions for the improvement of postoperative management of patients have demonstrated improvement in results. and the common denominator for each of these appears to be closer attention to the patient postope ratively. This factor alone will often be sufficient without any additions. mechani· cal or otherwise. Perhaps the success with in­ termittent positive pressure breathing on in­ spiration has been on the same basis.

Studies in the last decade tend to sup­ port this statement. In 1980, Schuppisser et al. found IPPB of no benefit over chest physiotherapy in altering ventilatory function or the incidence of pulmonary complications following upper abdomi­ nal surgery. Although both types of ther­ apy were equally effective, the authors recommend physiotherapy because of the added complications and cost associ­ ated with IPPB. The specific treatment referred to as "chest physiotherapy" is not defined in this study. Ali and co­ workers (1984) compared the added ef­ fect of IPPB with chest physiotherapy alone in 30 patients u ndergoing cholecys­ tectomy. Physiotherapy was described as deep breathing, coughing, turning, leg exercises, and early ambulation. The only statistical difference between the two groups was a severely depressed postoperative vital capacity in patients

293

ADJUNCTS TO CHEST PHYSIOTHERAPY

during IPPB treatment. The authors con­ clude that IPPB is very costly and not beneficial to patients receiving physio­ therapy after surgery. Although both studies reached similar conclusions, nei­ ther compared IPPB and chest physio­ therapy with a control group, in part for ethical reasons. A few studies have compared the effect of IPPB with incentive spirometry (IS) for reducing pulmonary complications after surgery and have failed to show an ad­ vantage of either therapy (Gale and San­ ders, 1980; Jung et aI., 1980; Alexander et aI., 1981 ). Similarly, Indihar and Associ­ ates (1982) found IPPB, IS, and turning with deep breathing and coughing to be equally effective in reducing pulmonary complications in 300 surgical patients. The variety of operative sites may have masked any possible benefit since ex­ tremity and lower abdominal surgery are less likely to cause pulmonary problems from secretion retention. Celli and co­ workers (1984) are one of the few inves­ tigators to use a control group when com­ paring the effect of IPPB, IS, and deep breathing exercises on preventing pul­ monary complications after abdominal surgery. They found all three methods to be significantly better than no treatment at reducing postoperative lung pathology; undesirable side effects were observed only in the 18% of patients receiving IPPB. Complications Associated with IPPB The complications associated with IPPB are shown in Table 9.1. Most are not life-threatening and were discussed ear­ lier in this chapter. Others have more se­ rious sequelae and require further discussion. IPPB is reported to cause broncho­ spasm which may produce increased air­ way obstruction and air trapping (Zi­ ment, 1973; Petty; Moore et a1.; Shapiro et aI., 1982). Karetzky (1975) reported acute pneumothorax associated with IPPB re­ sulting in fatality. These complications are believed to be particularly important when IPPB is used in the treatment of pa­ tients with acute asthma. Though pres­ sures commonly used during IPPB range between 10 and 1 5 cm H20, this limit is

Table 9.1 Reported Complications and Hazards of IPPB' 1 . Bronchospasm, distension of cysts and bullae, air trapping, pneumothorax 2. Gastric distension and ileus 3. Overdosage or adverse reactions of nebulized drugs 4. Acquired infection from contaminated equipment 5. Reduction of venous return causing hypotension 6. Reduction of respiratory drive in patients with chronic respiratory failure 7. Further impaction of mucus in patients unable or unwilling to expectorate following IPPB treatment 8. Exhaustion of patients who do not participate with treatment effectively 9. Complications of malfunctioning equipment 1 0 . Psychological dependence 'Adapted from Ziment (1 973), Karetsky (1975), and Shapiro et al. (1 982).

often exceeded both intentionally and unintentionally. This occurs when pa­ tients exhibit air trapping or are out of phase with the machine. Macklin and Macklin (1944) reported that intraalveo­ lar and pulmonary vascular pressure gra­ dients need not exceed 40 cm H20 before air can escape from the alveoli into the vascular sheaths. Since 1960, extrapul­ monary air has been reported with in­ creasing frequency in children with acute asthma (McGovern et aI., 1961; Jor­ gensen et aI., 1963; Bierman, 1967). This complication parallels the increased use of IPPB therapy in the treatment of child­ hood asthma. The pressurized flow of gas from the IPPB machine was also noted to cause gastric insufflation in patients. Re­ portedly, this can lead to colonic ileus and cecal perforation but more often re­ sulted in patient discomfort (Ruben et aI., 1961; Golden and Chandler, 1975; Gold, 1976). Humidification and medications are delivered in aerosol form by IPPB. How­ ever, IPPB ofers no advantages over other less expensive methods of delivering hu­ midity or aerosols (Gold, 1975; O'Dono­ hue, 1982; American Thoracic SOCiety). As with all nebulizers, IPPB nebulization is a source of bacterial contamination,

294

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

and careful sterilization of the apparatus is mandatory. Based on the spread of in­ fection attributed to IPPB, both Browner and Powers and Gold (1976) recommend abandoning such treatment for surgical patients. Because IPPB is most often pre­ scribed for persons prone to pulmonary complications, its ability to spread infec­ tion is significant. Sanders et al. (1970) re­ ported an outbreak of nosocomial infec­ tion with Serratia marcescens that was traced to contaminated medications de­ livered by IPPB. The number of patients with Serratia isolated from their sputum was proportional to the total number of IPPB treatments administered. Mertz and co-workers (1967) described an outbreak of Klebsiella pneumonia (resulting in five deaths in 1 month). All patients involved in the o utbreak received IPPB with bron­ chodilator treatments. The spread of this hospital-acquired infection was attrib­ uted to a contaminated stock bottle of bronchodilator solution. Contamination with gram-negative bacilli was reported as high as 91% in IPPB machines with reservoir nebulizers (Reinarz et al.). The Cost of IPpe IPPB is primarily used on surgical pa­ tients and those with COPD. The average number of treatments performed, aver­ age cost of treatment. and the types of pa­ tients receiving treatment differ from fa­ cility to faci lity. In 1974, McConnell and associates found the incidence of IPPB treatment at UCLA Hospital (700 beds) to be 7.000/year at a patient cost of $370,000/year. If these findings are ex­ trapolated to acute general hospital usage nationwide, an amount in excess of $400 million is figured. This sum, presumably, does not include amounts spent on out­ patient, extended care, or home care treatments and is based on 1974 costs. For the same year, Leith states that $2 billion/year was collected from the pub­ lic for IPPB treatments. In sampling both university and com­ munity hospitals, Baker (1974) found that a wide variation existed in IPPB usage (0.9-9% of all hospital admissions). Sim­ ilarly, in surveying the 43 Washington. D.C. hospitals that provide respiratory therapy, Schemer and Delaney reported

the average number of patients receiving IPPB per number of patient admissions to be 6% (range, 3-12%). Baker reported that generally, a greater number of treat­ ments were performed at hospitals asso­ ciated with schools of inhalation therapy and university hospitals showed a greater use of chest physiotherapy, sug­ gesting the application of alternative methods to IPPB treatments. Of interest, Schemer and Delaney also found that teaching hospitals showed a dramatic re­ duction (70%) in the number of IPPB treatments per 100 admissions between 1976 and 1979. Bet ween 1971 and 1979, the monthly number of IPPB treatments at Massachusetts General Hospital fell from a high of nearly 1 7,000 (1972) to a low of 500 (1 979), while the number of chest physiotherapy treatments re­ mained unchanged over this period (Pon­ toppidan). Braun et al. (1981 ) also re­ ported a significant decrease in IPPB use by two Wisconsin hospitals between 1971 and 1979. The hospital (university) hav­ ing a larger critical care population and generally more seriously ill patients showed a concurrent increase in the use of chest physiotherapy. The other hospi­ tal (a smaller private hospital) demon­ strated a trend of substituting incentive spirometry treatment for IPPB therapy. Pontoppidan attributes the rapid decline in IPPB therapy over the past years to increased challenges as to its cost effec­ tiveness, efficacy and scientific basis for use in both surgical and medical patients. In 1974. both usage and charges for IPPB showed wide variations that were not necessarily geographical in nature. The costs ranged from $3.75 to $7.50/ treatment; daily maximums ranged be­ tween $ 1 5 and $96 within the same city (Baker). In 1980, Hughes sampled five Chicago hospitals and found that initial treatment costs for IPPB ranged from $8.33 to $18. For all hospitals studied by Schemer and Delaney, an average of 60 treatments were given for every 100 pa­ tient admissions. However. large discrep­ ancies were noted between "for profit" hospitals, which averaged 190 treat­ ments/100 admissions, an'd federal hos­ pitals, where an average of 11 treat­ ments/100 admissions was performed.

ADJUNCTS TO CHEST PHYSIOTHERAPY

Likewise, the hospital bed number was • found to be Inversely related to the proportion of patients receiving lPPB. These findings are mostly supported by 0'00nohue (1985), who found significantly less use of lPPB for treating postoperative atelectasis in hospitals with more than 400 beds, compared to smaller facilities. He also reported a significantly lower use of chest physiotherapy in hospitals with 200 beds or less. Although lPPB appears to be more selectively prescribed from this study than in the past [Baker), it is still reportedly used in 82% of all hospi­ tals to treat postoperative atelectasis. This seems incongruous with the fact that studies have not proven that lPPB benefits surgical patients postoperatively [Gold, 1982). Another complication associated with lPPB is the use of contract services to provide it. To quote Petty, the use of this "practice is deplorable and offers the greatest chance of harm because the overuse and misuse of IPPB is very likely when there is no medical director as an established member of the hospital staff I where the 'service' is provided." The controls surrounding "big business" are a poor means of managing therapeutic treatment and do not lend themselves to minimizing patient costs. Respiratory therapy departments are reported to de­ pend on IPPB income for greater than 75% of their billing [Kittredge, 1973). A recent study in New England demon­ strates that IPPB usage can be reduced by greater than 92% without any change in mortality or morbidity [Zibrak et aI., 1986). Considering the investments in machinery, equipment, and payroll of contract companies, it is easy to imagine how IPPB became economically oriented rather than a carefully controlled aspect of patient care. Summary It is hard to conclude much that is ther­ apeutic about lPPB. Though it is an effec­ tive method of aerosol delivery, IPPB is more expensive yet no more effective than simple mechanically powered or hand-held nebulizers. Substantial con­ troversy exists over the ability of IPPB to improve alveolar ventilation, tidal vol-

295

ume, and arterial blood gases. At best, any improvements in pulmonary func­ tion accredited to IPPB are short-lived, since most treatment times are 20 min or less and are commonly given only 2-4 times/day. The psychological aspect of this treatment seems to vary consider­ ably and often depends on what other therapies are concurrently employed. IPPB is not an effective means of decreas­ ing or preventing pulmonary complica­ tions in surgical patients. Though the hazards associated with IPPB are rela­ tively small, the cost of IPPB is astronom­ ical. In addition, the relationship be­ tween contract services and both the quality and cost of their "service" should not be ignored. The ineffectiveness of IPPB may, in part, be attributed to the ab­ sence of uniform treatment indications and expectations. There must be clear ra­ tionale for implementing IPPB, along with a clear understanding of its indica­ tions, contra indications, risks, cost, and cost effectiveness [Petty). This has not oc­ curred in the past. The burden of proof of efficacy and economic justification lies in the hands of those who order and per­ form such treatment [Gold, 1975). The authors do not recommend IPPB for pro­ phylactic use or for the management of postoperative pulmonary complications. BRONCHODlLATlNG AND MUCOLYTIC AEROSOLS

A symposium of those that advocate the use of theophylline in the manage­ ment of COPD patients was published as a supplement to Chest [Vol. 92, 1S-43S, 1987). Ziment (1987) analyzes evidence for the effects of theophylline on muco­ ciliary clearance and suggests the drug may directly and indirectly improve mu­ cociliary clearance. Theophylline in­ creases the secretory output of bronchial glands. The transepithelial secretion of fluid into the respiratory tract is in­ creased. Theophylline stimulates the chloride pump that is controlled by cy­ clic AMP. Ciliary motility is also stimu­ lated by theophylline. The major effects of theophylline that are likely to be ben­ eficial in the CO PO patient occur proba­ bly due to bronchodilation and because theophylline improves diaphragm con-

296

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

tractility and reverses diaphragm fatigue (Aubier. 1987). Whether the bronchodi­ lating effects of theophylline occur by aerosol inhalation alone or are improved by intravenous administration and addi­ tion of f/,-agonists such as salbutamol re­ mains controversial (Brain. 1984; jenne. 1987a). There is no evidence that theoph­ ylline is beneficial in the patient with acute lung disease. Benefit from use of aerosol-delivered mucolytic agents is poorly substantiated. with the possible exception of its use in children with cystic fibrosis (Brain. 1984). There are no data to substantiate the use of mucolytic agents to assist removal of secretions in acute lung disease. In our clinical experience. neither bronchodi­ lators or mucolytic agents delivered by aerosol appeared to be of benefit in pa­ tients with acute lung disease receiving chest physiotherapy. Intuitively it seems that benefit for acutely atelectatic lung is likely to be minimal since nonventilated areas of lung do not directly receive any aerosolized drugs. Intravenously admin­ istered bronchodilators. given by contin­ uous infusion after a loading dose and monitored by blood levels. appear bene­ ficial in reversing severe bronchocon­ striction. Mild bronchospasm is fre­ q uently found to reverse following the removal of retained secretions. which is assisted by adequate systemic hydration of the patient. inhaled humidity and chest physiotherapy. However. the use of aerosol-delivered mucolytic agents and bronchodilators is widespread in the management of pa­ tients with acute chest disorders. It ap­ pears that these therapies are inappropri­ ately and. often. reflexly prescribed. A review of the literature does not provide many concrete facts about the indica­ tions for and clinical effectiveness of these aerosols (Brain and Valberg. 1979). Bronchodilators There is evidence that some broncho­ dilating aerosols are useful in certain pa­ tients with chronic pulmonary disease. particularly asthma and cystic fibrosis. Beta-adrenergic aerosols tend to have a rapid and predictable onset with peak ef-

fects detected in minutes. but the efficacy of aerosol compared to intravenous use of bronchodilators needs to be deter­ mined (Brain. 1980; jenne. 1987a). The indications for use and toxicity of bron­ chodilators in young children. in the presence of cardiovascular impairment. and in the long-term therapy of such dis­ eases as COPD. asthma. and cystic fibro­ sis need to be determined (Featherby et al.. 1970; Brain. 1980; jenne. 1987b). The significance of the nonbronchodilator pharmacological effects of these medica­ tions also requires further investigation (Brain. 1980). The benefits of using bron­ chodilators in patients with acute lung pathology including lung contusion and flail chest are not established. In this patient population. some clinical signs of bronchospasm. such as wheeZing. may be relieved by removing excess secretions. Mucolytic Agents Mucolytic agents are ineffective in im­ proving airway clearance. with the pos­ sible exception of patients with cystic fi- . brosis (Thomson et al.. 1975; Brain and Valberg; Brain. 1980. 1984). Although aerosols of these drugs can alter mucus characteristics in vitro and probably in vi vo. it is not known if the changes en­ hance mucociliary activity and the cough mechanism (Brain and Valberg). Ade­ quate humidification (100% relative hu­ midity) is probably the best means of Iiq­ uifying secrections and is without the side effects associated with mucolytic agents (Wanner and Rao. 1980). Bland aerosols including distilled water. saline. and half normal saline have no substan­ tiated benefits in treating lower airway disease. Moreover such aerosols may elicit bronchoconstriction in adults and children (Brain. 1984). Present understanding of aerosol de­ position and location of airway receptors for various drugs is inadequate (New­ house and Ruffin. 1978). For aerosol de­ livery to be effective. particle size must be tailored to the delivery system and site of receptors as well as coordinated with specific breathing maneuvers. Posi­ tive pressure breathing of aerosols does not produce a more uniform or more pe-

ADJUNCTS TO CHEST PHYSIOTHERAPY

ripheral distribution of aerosols than quiet breathing (Newhouse and Ruffin. 1978). Complications It is difficult to measure the received dose of a drug given by aerosol (Brain and Valberg). Even if the principles govern­ ing aerosol deposition. retention. and clearance are understood. the amount of medication reaching a specific area of the lungs is only an estimate (Brain. 1980). Although the small particles produced by some nebulizers are capable of reaching the terminal airways. the percentage of the drug delivered to these areas also de­ pends upon the tidal volume. breathing frequency. expiratory reserve volume and length of breath holding of the pa­ tient. the presence of an artificial airway. and the length of delivery tubing from the source to the patient (Brain and Val­ berg; Swift. 1980). Nose breathing com­ pared with mouth breathing effects par­ ticle deposition and can markedly increase the quantity of medication re­ tained by the nasal m ucosa and pharynx (Brain and Valberg). This. in turn. may be swallowed and systemically absorbed. resulting in general effects as opposed to local changes. The efficacy of a broncho­ dilator or mucolytic agent on the lungs depends on successful delivery to the af­ fected area. In the presence of pulmonary collapse. obstruction. or constriction. air­ flow and. therefore. aerosol flow are im­ peded. Nonventilated areas of the lung do not directly receive any benefit from inhaled medications (Brain. 1980). Many aerosol devices used by patients with chronic lung diseases require intelligent use by the patient and they are fre­ quently misused (Brain. 1984). The possibility of infection through contamination by aerosol-generating equipment is well documented in the lit­ erature and is discussed on pp. 284 and 293. Because aerosols can be deposited in the most distal airways and are most often used on patients already suffering from pulmonary complications. the added insult of iatrogenic i nfection can be significant and should be avoided by proper sterilization. In mechanically ventilated patients re-

297

celvlng chest physiotherapy. nebuliza­ tion of N-actey1cysteine (a mucolytic agent) rendered subsequent physiother­ apy treatments less effective in increas­ ing pulmonary compliance (Winning et al.. 1975). N-Acety1cysteine is also re­ ported to cause bronchoconstriction in patients with asthma and chronic airway disease (Bernstein and Ausdenmore. 1964; Rao et al.. 1970) and in normal in­ tubated subjects (Waltemath and Berg­ man. 1973). It appears that based on the cost and potential hazards of mucolytic agents and bronchodilators. use should be re­ stricted to those aerosols whose clinical value is documented. The currently pre­ scribed mucolytic aerosols do not meet this requirement (Wanner and Rao. 1980). The indications. complications. and benefits of aerosol-delivered bron­ chodilators in patients suffering acute lung pathology are not established (Brain. 1980). The minimal effective dose. opti­ mal method of delivery. and type of patient who can most benefit from short­ term and continued bronchodilator treat­ ment still need to be determined (Brain. 1980).

MECHANICAL DEVICES USED TO ENCOURAGE LUNG EXPANSION FOLLOWING SURGERY

For many years. mechanical aids to lung expansion enjoyed tremendous pop­ ularity without critical analysis of either their rationale or usefulness (Pontoppi­ dan). A wide array of maneuvers and de­ vices were suggested in attempts to pre­ vent pulmonary complications after surgery. Transtracheal aspiration and deep breathing were proposed. and these are discussed elsewhere (see pp. 164 and 119). Inhaling carbon dioxide. once be­ lieved to be of benefit (Alder. 1967; Jones. 1968). is no longer used. Aside from IPPB and breathing exercises. the methods most in vogue appear to be forms of in­ centive spirometry (IS). continuous posi­ tive airway pressure (CPAP). and posi­ tive expiratory pressure (PEP). Blow bottles represent a type of expiratory IS. while most currently used IS emphasize the inspiratory phase of respiration. For

298

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

clarity, the term IS refers only to the in­ spiratory devices in this text. Both CPAP and PEP primarily affect expiration. The use of these techniques is based on research by Anderes et a!. (1979), who found improved lung reexpansion through collateral airway channels when CPAP was applied. Blow Bottles In eight patients, Colgan et a!. (1970) studied the effects of resistance breathing on FRC, using blow bottles and sustained hyperintlations with the Elder demand valve resuscitator. Both methods were found to produce marked increases in airway pressure similar to that seen dur­ ing a Valsalva maneuver, and neither was found to be beneficial in treating at­ electasis following surgery. Significant increases in FRC were reported after blow bottle use, but this was believed to be the result of the sustained deep breath performed prior to treatment. O'Connor (1975) also examined blow bottles, but he compared them to a device that added in­ creased dead space and expiratory pres­ sure. In the 23 patients studied following laparotomy, a significantly greater in­ crease in vital capacity was reported in those using the dead space, expiratory pressure device. Based on these findings, O'Connor suggested that this device may result in a decreased incidence of respi­ ratory complications after surgery. Fur­ ther information on this device is not available. Heisterberg et a!. (1979) com­ pared the effect of using blow bottles (for 10 min every 4 hr) with chest physiother­ apy (breathing exercises, postural drain­ age, and coughing, two times per day) in 98 patients u ndergoing elective gastric or biliary tract surgery. They found the incidence of radiological pulmonary changes to be the same in both groups and concluded that blow bottles are pref­ erable to chest physiotherapy because they are less time consuming. There are no documented complications of blow bottle use; theoretical concerns include hyperventilation, increased atelectasis, barotrauma, and cost (Shapiro et a1.. 1982). Based on the existing literature, blow bottles have not been found more effective than other techniques and have

not been shown to decrease pulmonary complications postoperatively. Adjuncts emphasizing inspiration, not expiration, are thought to be better at improving res­ piratory function after surgery. Incentive Spirometry Many authors compared IS to IPPB as a means of reducing respiratory complica­ tions after surgery (see p. 293). Both of these devices emphasize inspiration; IPPB is performed passively, while in­ centive spirometry involves an active maneuver toward maximal inspiration. Van de Water and associates [1972) stud­ ied 30 consecutive women following ad­ renalectomy; 15 of them received IS and the rest IPPB. Both groups were able to be treated with other therapies, which in­ cluded blow bottles, rebreather tubes, and thoracic physiotherapy. Only three patients in the IS group developed py­ rexia (at least 38.5·C) and clinical find­ ings indicative of pulmonary complica­ tions, compared with six of the patients who received IPPB. The average hospital stay was 9 days for those treated with IS and 1 1 days for IPPB-treated patients. The authors claim good patient accep­ tance of IS at one-tenth the cost of IPPB. Due to the variety of therapies used in each group, it is not possible to tell which therapy, if any, actually altered the morbidity. Iverson and co-workers [1978) evalu­ ated three methods commonly used after surgery to reduce pulmonary complica­ tions. Of the 145 patients in their study, 42 received IPPB, 45 used blow bottles, and 58 received IS. The incidence of pul­ monary complications, evaluated by chest x-ray interpretation, clinical find­ ings, and arterial blood gas results, was 30% with IPPB, 8% with blow bottles, and 15% with IS. Also associated with IPPB was a significant increase in gastro­ intestinal complications. Dohi and Gold (1978) studied 64 patients: 30 received IPPB and the rest received IS. All patients were treated for 5 days after surgery and were observed for pulmonary complica­ tions by chest x-ray and clinical exami­ nation. Because the data favoring IS over IPPB were statistically slim, the authors claimed no conclusive difference be-

ADJUNCTS TO CHEST PHYSIOTHERAPY

tween the two therapies. Based on these findings, they also stated that the use of IS rather than IPPB may be justified, since the former is much cheaper. An­ other study comparing the postoperative use of IS and IPPB did not find a change in the incidence of pulmonary dysfunc­ tion with the two modes of therapy (Gale and Sanders, 19S0). The previous authors failed to compare either IPPB or IS with deep breathing. However, McConnell et al. contrasted the effects of voluntary deep breathing, IPPB, and IS on transpulmonary pressure gra­ dients in 1 1 thoracotomy patients and 6 normal subjects. Deep breathing pro­ duced an average gradient of 24.6 cm H20, IPPB of 2 1 . 7 cm H20, and IS of 29.4 cm H20. The authors state that increased transpulmonary pressure gradients are a principal determinant of alveolar and bronchial expansion. Therefore, they claim that IS is more convenient and less costly than IPPB in achieving alveolar and bronchial expansion. The fact that deep breathing is even less expensive arid requires no more instruction than the other two methods was not ad­ dressed. Alexander and co-workers (19S1) failed to show a decrease in pul­ monary complications when IS (up to So% of the preoperative maximal inspir­ atory volume), IPPB (3 times per day), or IS with IPPB was compared to a control group (encouraged to breathe deeply and ambulate). Similarly, Indihar et al. (19S2) found no benefit of IS or IPPB over turn­ ing, coughing and deep breathing in 100 surgical patients. Two studies compared the effects of IS and continuous positive airway pressure (CPAP). One evaluated CPAP, IS, and coughing and deep breath­ ing in 65 adults. Patients received their assigned treatment for 15 min, every 2 hr (while awake), for 3 days after abdominal surgery (Stock et aI., 19S5). The authors concluded that IS offered no advantage over coughing and deep breathing. Rick­ sten et al. (19S6) studied 43 similar pa­ tients but compared IS to CPAP and PEP. They found both CPAP and PEP superior to IS in improving gas exchange, lung volumes, and radiological clearing after upper abdominal surgery. A study comparing three types of IS used following upper abdominal surgery

299

was undertaken i n 79 patients (Lederer et aI., 19S0). Instruction in how to use the assigned device was given prior to sur­ gery. Though only monitored once daily afterwards, all patients were encouraged to use their assigned IS 10 times every waking hour. On each day following sur­ gery, a substantial number of patients in each group did not use their device at all. Other types of therapy, including ultra­ sonic nebulization, chest percussion, postural drainage, or any combination of these, were given to some patients in each group. There was liltle statistical dif­ ference between the three types of IS in terms of the patient's pulmonary func­ tion, vital signs, and white blood cell count, and there was no difference in the length of hospital stay. It would be more interesting if the three groups were com­ pared to both a control group and a group performing voluntary deep breathing. Because of the lack of a control group, i t i s still not known whether any device at all was of benefit to these patients (Hughes, 19S0). A few studies evaluating the use of IS were performed on patients requiring cardiac surgery. Krastins et al. (19S2) studied 17 children and found that when IS (every 2 hr for 12 hr a day) was used in conjunction with chest physiotherapy, a dramatic decrease in atelectasis resulted. However, there was no difference be­ tween the postoperative pulmonary function tests (PFT) of children receiving IS with physiotherapy (study group) and those receiving chest physiotherapy (control group). The incidence of atelec­ tasis in this study was higher than that reported by others (SS%); also pleural ef­ fusion occurred in all control and two study patients. In 25 adults undergoing coronary artery surgery, the efficacy of two different IS techniques was com­ pared with chest physiotherapy (cough, deep breathing, postural drainage, per­ cussion, and vibration) (Oulton et al.. 1 9 S 1 ). No added benefit was found when one type of IS (Triflow) was performed in addition to chest physiotherapy; how­ ever. when the other IS (Spirocare) was used. fever and less severe pulmonary complications were noted on chest x-ray. The investigators attributed the differ­ ence between IS modalities to the fact

300

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

that Triflow is more flow dependent while Spirocare is more volume sensi­ tive. These conclusions are in conflict with those of Lederer et a!. who found no significant difference between the two IS studied by Oulton et a!. or another IS (Bartlett-Edwards) that is also volume de­ pendent. Additionally, the Spirocare unit is more expensive than other IS, is not disposable, and requires an electrical power source (Van de Water, 1980). Vraciu and Vraciu (1977) studied the effects of physical therapist-assisted breathing exercises that included lateral and posterior basal expansion, diaphrag­ matic breathing, and coughing on 40 pa­ tients following open heart surgery. The patients who performed breathing exer­ cises had statistically fewer pulmonary complications (3 of 19) than the controls (8 of 2 1 ), even though both groups re­ ceived IS every 2 hr. The control patients were also turned hourly and assisted in deep breathing and coughing by the nursing staff. Pulmonary complications were defined by the presence of a tem­ perature greater than 38.S'C, radiological evidence of atelectasis, and abnormal breath sounds. In 1983, Dull and Dull compared the ef­ fects of early mobilization to deep breathing and IS in 49 patients after car­ diopulmonary bypass surgery. Mobiliza­ tion consisted of extremity exercises, coughing, and assistance with turning, sitting, or standing. Because IS could not be performed during mechanical venti­ lation, the patients began their assigned exercise regimen within 4 hr of extuba­ tion. The incidence of pulmonary com­ plications was high in this study and was defined as a temperature increase of 4'F above baseline or 2-3'F increase in tem­ perature along with rales, rhonchi, ab­ sent breath sounds, or purulent sputum. The authors conclude that neither IS nor deep breathing offers any therapeutic ad­ vantage over patient mobilzation in pre­ venting pulmonary complications after cardiopulmonary bypass surgery. Craven et a!. (1974) compared the ef­ fects of chest physiotherapy (including postural drainage, percussion, breathing exercises, and assisted coughing) to IS in 70 patients after surgery. Increased pul-

monary complications were found in both groups of patients having prior chronic respiratory disease, although pa­ tients receiving IS showed significantly fewer complications than those receiving chest physiotherapy. The authors con­ cluded that though there is a need for chest physiotherapy in treating major pulmonary collapse, IS may be a more ef­ fective means of prophylaxis. Work by other investigators (Vraciu and Vraciu, 1977; Lyager et aI., 1979; Van de Water, 1 980) appears to conflict with that of Craven et a!. (1974). Lyager and associates studied the added effect of IS (every waking hour) on patients already receiving instruction by physical therap­ ists. On the basis of random selection, 34 upper abdominal surgical patients were given physical therapist-assisted breath­ ing exercises and coughing, and 49 re­ ceived IS in addition to this treatment. Radiological and clinical findings (in­ cluding coughing, expectoration, dys­ pnea, auscultation, temperature, and res­ piratory rate) and arterial blood gas analysis revealed no differences between the two groups. The incidence of,pulmo­ nary complications after surgery was also the same, even in the subgroup of pa­ tients who used their IS on the average of 10 times every hour. The results of this study suggest that no further benefit is gained by adding IS to aggressive pul­ monary physiotherapy. Although the re­ sults are of doubtful clinical significance, Hedstrand and associates (1978) also found that physical therapist-encouraged techniques of deep breathing were supe­ rior to voluntary or device-assisted deep breathing in raising arterial oxygenation. A controlled study was undertaken to as­ sess whether bedside coaching to breathe deeply was as effective as IS in prevent­ ing postoperative atelectasis (Van de Water, 1980). Encouragement by the nursing staff was superior to IS in return­ ing pulmonary function to preoperative values. The authors conclude that fre­ quent patient contact appears to be an important component. This human ele­ ment opens up an area of subjective dif­ ferences. The rapport between an indi­ vidual therapist or nurse and patient may well make a difference in patient perfor-

ADJUNCTS TO CHEST PHYSIOTHERAPY

mance. Likewise, the emphasis placed by a therapist on a maneuver or treatment may also influence its effectiveness. In 1986, Schwieger et al. performed a well-designed, controlled study on 40 pa­ tients to assess the effect of IS following cholecystectomy. IS use was supervised and consisted of a slow, deep inspiratory effort with a volume oriented device. IS was performed for 5 min, hourly, at least 12 times a day, for 3 days after operation. All patients were mobilized on the day of surgery. Subjective and objective clinical data (oxygenation, temperature, PFT, white blood cell count, chest x-ray, and auscultation) were evaluated to establish the incidence of pulmonary complica­ tions. There were no significant differ­ ences in any of the measured parameters between the two groups, although chest radiograph changes indicating lung pa­ thology occurred in 40% of the IS patients compared with 30% of the controls. None of the study subjects was classified as high risk for developing postoperative respiratory problems. This study does not support the theory of some investi­ gators who attribute the lack of benefit from IS in controlled studies to inade­ quate supervision, coaching, or patient use. Although compelling support for pro­ phylactic or paliative IS therapy is not available, few complications are associ­ ated with its use. Theoretical problems include hyperventilation and baro­ trauma (Shapiro et al.). Cost is also an issue, especially when more expensive or nondisposable types of IS are recom­ mended. IS is reportedly used in 95% of the hospitals surveyed in the United States as a prophylactic maneuver to im­ prove lung expansion and in the treat­ ment of postoperative atelectasis (0'00nohue, 1985). This compares with a use of 44% in the United Kingdom (in high­ risk, postoperative coronary artery by­ pass patients) (Jenkins and Soutar, 1 986). Cost and questionable efficacy were the major reasons for the much lower use of IS in Great Britain. Even a relatively small cost becomes significant when multiplied by 95% of hospitalized pa­ tients receiving surgery in the United States. It is the authors conclusion that

301

after surgery, early mobilization and vo­ litional deep breathing offer more cost-ef­ fective treatment than IS for the sponta­ neously breathing patient at risk for postoperative pulmonary complications. Continuous Positive Airway Pressure and Positive Expiratory Pressure Although originally used with neo­ nates, face-mask CPAP is recommended by some as a means to reduce or reverse the incidence of postoperative respira­ tory complications in adults. The use of PEP, a modification of CPAP, has also been studied. In 1 979, Anderes and co­ workers evaluated the effect of CPAP and positive and expiratory pressure (PEEP) on 30 adults undergoing elective upper abdominal surgery. Half of the pa­ tients [Group A) were ventilated without PEEP during surgery, were extubated, and breathed spontaneously afterward. The others [Group B) received PEEP (10 cm H,O) during anesthesia and CPAP [3 cm H,O) while intubated for 3 hr before extubation. The author reported signifi­ cant deterioration in PaO, and right to left shunt [Q./QT) and adverse radiologi­ cal findings in Group A compared with Group B. These changes occurred over 3 days after surgery although PEEP was given only during, and CPAP was given once, only immediately after surgery. I t i s difficult t o separate t h e effects o f CPAP from PEEP in this study. Carlsson and as­ sociates (1981) studied 24 patients under­ going elective cholecystectomy. Face­ mask CPAP (about 5-10 cm H,O) was given to 13 patients for 4 hr after surgery; the other patients wore a face mask but did not receive CPAP. PEEP was not used. All subjects were evaluated during treatment and for 24 hr postoperatively. No significant difference between groups was found in chest x-ray findings or spi­ rometry or blood gas measurements. The results of these two investigations are dif­ ferent although similar patients were studied. Information on other types of pulmonary therapy, including patient mobilization, was not addressed but may be responsible for the different findings. Anderson et al. (1 980) evaluated the ef­ fect of face-mask CPAP in reversing at-

302

CHEST PHYSIOTHERAPY I N THE INTENSIVE CARE UNIT

electasis in a variety of surgical patients during the first 24 hr after operation. Twelve subjects received conventional therapy (including posteral drainage, deep breathing, and suctioning three times a day) and 12 patients additionally received CPAP (at about 15 cm H,O for 25-35 respirations, hourly while awake and twice at night). Treatment was con­ sidered successful if PaO, increased by 1 5% of predicted value or if chest x-ray findings improved by 50%. CPAP was found to significantly improve atelectasis resolution; augmented collateral ventila­ tion was thought to be the mechanism re­ sponsible for the change. This is one of the few studies that used a modality to reverse, rather than prevent, chest pathology. Two studies compared the effects of IS and CPAP in preventing postoperative pulmonary complications. In 65 adults u ndergoing elective upper abdominal surgery, Stock and associates (1985) eval­ uated the use of face-mask CPAP, IS, and coughing with deep breathing. All treat­ ments were given for 15 min, every 3 hr while awake for 3 postoperative days. Very few differences were found be­ tween treatment groups; a more rapid in­ crease in FRC occurred in patients who received CPAP, but x-ray changes and fever were not significantly different be­ tween groups. In 50 patients undergoing similar surgery, Ricksten et al. (1 986) compared IS, CPAP (10- 1 5 cm H,O), and PEP (10-15 cm H,O). Each treatment was applied for 30 breaths, every waking hour, for 3 days after operation. The au­ thors reported that both CPAP and PEP were superior to IS with respect to im­ proving gas exchange, preserving lung volumes, and resolving atelectasis after upper abdominal surgery. They advocate using the simple PEP mask since it was as effective and less complicated than face-mask CPAP. Conflicting results are also found on the effect of PEP on patients with cystic fibrosis. Falk et al. (1984) reported that PEP i ncreased skin oxygen tension and sputum production over that obtained from posteral drainage with percussion and vibration or forced expiratory tech­ nique (FET). No radiographic changes

were observed in any study situations. Contrary findings are reported by Hof­ meyr and co-workers (1986), who found that oxygen saturation was unchanged and that sputum clearance was less ef­ fective when PEP (12-17 cm H20) was incorporated in a regimen of breath­ ing exercises, postural drainage, and FET. Two studies have looked at whether intermittent PEP augments the effects of chest physiotherapy in the management of postoperative pulmonary complica­ tions. Campbell and associates (1 986) randomly assigned 71 patients to receive chest physiotherapy (control) or chest physiotherapy plus PEP (study group) after abdominal surgery. Chest physio­ therapy in both groups consisted of breathing exercises, huffing, and cough­ ing, while sitting, every 2 hr. Postural drainage was used if indicated. The study group also received PEP for 10 breaths every 2 hr. Thirty-one percent of the con­ trol patients developed respiratory com­ plications compared to 22% of the study patients. This difference corresponded to the larger number of smokers in the con­ trol group. The effect of PEP in addition to chest physiotherapy was studied in 56 patients after thoracotomy (Frolund and Madsen, 1986). Chest physiotherapy was given two times a day and consisted of early mobilization, arm exercises, deep breathing, and coughing. Patients were encouraged to use face-mask PEP (about 10 em H,O) at least 10 m'n each waking hour and were supervised in its use two times a day. All patients were evaluated for 3 days after surgery. The authors con­ clude that PEP offered no benefit over conventional physiotherapy in prevent­ ing atelectasis or improving arterial oxygenation. Although research on the effects of postoperative PEP or CPAP is conflicting, complications are not associated with its use. The added expense of this modality is a concern, but research on the cost of PEP or CPAP is scarce. In the treatment of postoperative atelectasis, CPAP is re­ portedly used in 25% of hospitals in the United States and use is dramatically greater in hospitals with more than 200 beds (O'Donohue, 1985).

ADJUNCTS TO CHEST PHYSIOTHERAPY

Summary Few conclusions can be drawn from all the research on the mechanical aids to lung expansion discussed in this section. Differing durations and frequencies are recommended for the same device, and few authors compared similar tech­ niques, other than contrasting lPPB to IS. It is no surprise that the results are so in­ consistent. Only the finding that [PPB is more costly and less effective than the other methods mentioned appears with regularity. The benefit (or lack of i t ) ob­ tained from using blow bottles is no more substantiated than the outdated method of having a patient inflate a rubber sur­ gical glove (Hl1ghes). Similarly, there is no supporting evidence that blow bottles decrease the incidence of pulmonary complications following surgery (Pontop­ pidan). Though forced expiration against resistance is encouraged by those rec­ ommending and ordering blow bottles (or inflatable surgical gloves), forced expira­ tion alone is reported to produce atelec­ tasis and hypoxemia (Nunn et aI., 1965). Pontoppidan notes that improper use of blow bottles may decrease both end-ex­ piratory volume (if the patient exhales too forcefully) and cardiac output due to the increased airway pressure. While in agreement with Kasik and Schilling (198 1 ) that little is gained by re­ placing [PPB with yet another form of therapy (such as IS, CPAP, PEP, or chest physiotherapy techniques) that is still under investigation, the ethical issue of withholding a treatment that has been shown to be beneficial by some, cannot be ignored. Methods that encourage max­ imal inspiration, such as [S and deep breathing, may, in theory, be more ben­ eficial than blow bottles. However, the available research has not proved [S su­ perior in terms of cost and effectiveness to deep breathing and the other aspects of chest physiotherapy. The principle that CPAP improves collateral ventilation and therefore lung reexpansion is excit­ ing. However, studies on the efficacy of CPAP or PEP to reduce postoperative pulmonary complications are conflicting. These modalities appear to be of little added benefit to a regimen of chest phys-

303

iotherapy including patient mobilization. It is important to note that CPAP, PEP, IS, blow bottles, and [PPB are of limited value in the [CU, since they cannot be used on patients requiring mechanical ventilation. Chest physiotherapy does not have this limitation. Despite the huge cost and widespread use of mechanical aids following surgery, the incidence of respiratory complica­ tions has not significantly changed with any single or combination of treatments. It appears that the type of patient who could perhaps benefit from IS, CPAP, or PEP still needs to be defined. Alternately, it may mean that none of the described modalities represents an optimal method of preventing pulmonary complications. Ford and Guenter (1984) suggest research into specific diaphragm function during the early postoperative phase (see p. 1 5 ). BRONCHOSCOPY

Bronchoscopy is used both therapeuti­ cally and diagnostically. The indications for therapeutic bronchoscopy include as­ piration, secretion retention, atelectasis, lung contusion, and lung abscess (Wan­ ner et aI., 1973; Lindholm et aI., 1974; Sackner, 1975; de Kock, 1977; Barrett, 1978; Dreisin et aI., 1978). Bronchoscopy is often used as an adjunct to chest phys­ iotherapy. However, since chest physio­ therapy was introduced to our facility in mid-1973, the need for therapeutic bron­ choscopy has diminished markedly (Table 1 .2). Complications, restrictions, and precautions of therapeutic bronchos­ copy are discussed and compared to chest physiotherapy. Complications and Precautions Suratt and co-workers (1976) reported a 0.022% death rate and a 2.92% inci­ dence of serious-to-life-threatening com­ plications associated with fiberoptic bronchoscopy. [n reviewing 24,521 fiber­ optic bronchoscopic procedures, Credle et al. (1973) reported a low incidence of major complications « 1 %) and a mortal­ ity of 0.01%. [n prospective studies, Dreisin et al. and Pereira et al. (1978)

CHEST PHYSIOTHERAPY IN THE INTENSIVE CARE UNIT

304

found the overall complication rate to be 1 1 % and 8%. repectively (Table 9.2). None of the authors reported the mortal­ ity or morbidity resulting solely from use of fiberoptic bronchoscopy in the treat­ ment of retained secretions. and a wide spectrum of patients ranging from those in the lCU to the office outpatient were included in these studies. Pneumonia. bronchospasm and laryngospasm. pneu­ mothorax. hypoxemia. hemodynamic changes. cardiac dysrhythmia. respira­ tory arrest. and hemorrhage of sufficient quantity to compromise the airway are among the serious complications associ­ ated with bronchoscopy (Credle et al.; Dubrawsky et al.. 1973; Harrell et al.. 1973; Britton and Nelson. 1974; Albertini et al.. 1974; Karetzky et al.. 1974; Salis­ bury et al.. 1 975; Feldman and Huber. 1976; Sahn and Scoggin. 1976; Suratt et

al.; de Kock; Dreisin et al.; Periera et al.. 1978; Shrader and Lakshminarayan. 1978; Lundgren et al.. 1982). In addition to pneumonia. febrile reac­ tions and bacteremia are noted following both rigid and fiberoptic bronchoscopy (Burman. 1960; Pereira et al.. 1974. 1978; Timms and Harrell. 1975; Dreisin et al.). Patients over 60 years of age or those with a history of cardiovascular disease or immunoincompetence appear to be at increased risk of developing these com­ plications. In patients with bronchial asthma. life-threatening laryngospasm and bronchospasm from fiberoptic bron­ choscopy are reported despite prior premedication. application of topical an­ esthetics. and supplemental oxygen de­ livery (Sahn and Scoggin; Dreisin et al.; Pereira et al.. 1 978). As a result. fiberoptic bronchoscopy should be performed with

Table 9.2 Reported Complications Within 24 hr of Fiberoptic Bronchoscopy (Prospective Studies)' Complications

Dreisin at al. (205 Procedures)

Major Pneumonia Bronchospasm/laryngospasm or airway obstruction Pneumothorax requiring chest tube Hemoptysis (40 ml in 1 5 min or 200 ml in 24 hr) Respiratory arrest Total Minor Vasovagal reactions Fever Cardiac dysrhythmias Bleeding (nosebleed) Obstruction of airways Infiltrates without fever Nausea and vomiting Pneumothorax Dyspnea Subcutaneous emphysema Electrocardiogram abnormality Acute maxillary sinusitis Psychotic/hysterical reaction Aphonia Total 'Adapted from Dreisin et al. (1 978) and Pereira et al. (1 978). '-Information not given.

Table 9.3 Reported Changes in Arterial Blood Gases from Fiberoptic Bronchoscopy' Author

Number and Type of Patient (A and B Denote Subgroups)

Reported Changes (Before to After, Unless Stated)

:;: Comments

Dubrawsky et al. (1 973)

49, SB A. 30, room air B. 19, supplemental 0,

PaCO, and pH unchanged A. PaO, ! 22.4 mm Hg (p < O.OOOS) after insertion PaO, ! 18.8 mm Hg (p < O.OOOS) after lavage and suctioning B. PaO, unchanged

60 to 80 ml of lavage used Bronchial lavage results in hypoxemia Initially hypoxemic and normal patients showed similar PaO, ! after procedure

Harrell et al. (1 973)

1 S, SB

PaO, ! > 1 a mm Hg (in 2 patients) PaCO, t ;":S mm Hg

Degree of hypoxemia did not correlate with extent of lavage or preprocedural PaO, levels Degree of hypoxemia did relate to procedure time and amount of suctioning t PaCO, was related to bronchospasm

Kleinholz et al. (1 973)

Albertini et al. (1 974)

10, SB

1 8, SB A. 1 6, SB B. 1 S, SB (FlO, = 1 .0)

Karetzky et al. (1 974)

1 4, SB

Pierson et al. (1 974)

1 0 MV

,

6 patients, PaO, ! ;,,: 9 mm Hg; 3 patients, PaO, unchanged; 1 patient, PaO, t from hyperventilation

No supplemental 0, given

A. PaO, ! 20 mm Hg (range, 4-38) B. A-aDO, t S6 mm Hg (range, 21 88)

! PaO, did not correlate with amount of anesthetic or lavage material used or duration of treatment PaO, remained ! from < 1 hr to >4 hr; most returned to values obtained before procedure by 2 hr

m

�

." I -< (f)

5 .... I

PaO, ! 1 2 mm Hg during (range, -2--21) PaO, ! S m m Hg after (range, +3- 1 9) PaO, t 1 0 mm Hg during (range, + 1 00- - 1 32) PaCO, t S.2 mm Hg during (range, -4-+ 14) pH ! 0.06 during PaO, t 14.3 mm Hg after (range, +7S--39)

() I

m :D » ." -< Z .... I

m Z ....

m

z (f)

All patients had retained secretions or atelectasis FlO, t to 1 .0, 20 min prior to procedure in most instances

22 mm Hg PaO, did not ! significantly with t treatment time (average, 25 min) PaO, returned to initial level 1 5-30 min ... after procedure No patients had significant amounts of secretions

» o 20 80

••

'0

.

,

• cpp

" "

� !

t

gam

,

� � e,

o

1 .. � •

,. 100m

I

0

�

0

0

,� "

"

� �

"

� /iff .. ..

t A }.. , 1 10 m